/*
* $RCSfile: Perceptron.java,v $
* $Revision: 1.11 $
* $Date: 2007/06/30 17:30:33 $
* $Author: wojna $
*
* Copyright (C) 2002 - 2007 Logic Group, Institute of Mathematics, Warsaw University
*
* This file is part of Rseslib.
*
* Rseslib 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.
*
* Rseslib 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 <http://www.gnu.org/licenses/>.
*/
package rseslib.structure.function.doubleval;
import rseslib.structure.data.DoubleData;
import rseslib.structure.data.DoubleDataWithDecision;
import rseslib.structure.table.DoubleDataTable;
/**
* Perceptron.
*
* @author Arkadiusz Wojna
*/
public class Perceptron implements DoubleFunction
{
/** Array of input functions. */
DoubleFunction[] m_arrComponents;
/** Weights assigned to input functions. */
double[] m_arrCoefficients;
/**
* Constructor.
*
* @param components Array of input function.
* @param coefficients Weights assigned to input functions.
*/
public Perceptron(DoubleFunction[] components, double[] coefficients)
{
m_arrComponents = components;
m_arrCoefficients = coefficients;
}
/**
* Returns weights assigned to input functions.
*
* @return Weights assigned to input functions.
*/
public double[] getWeights()
{
return m_arrCoefficients;
}
/**
* Returns the value of this function for a given double data.
*
* @param dObj Double data to be evaluated.
* @return Value of this function for a given double data.
*/
public double doubleVal(DoubleData dObj)
{
double d = 0;
for (int i = 0; i < m_arrComponents.length; i++) d += m_arrCoefficients[i]*m_arrComponents[i].doubleVal(dObj);
return d;
}
/**
* Returns an array of numeric attributes as the array of input functions.
*
* @param size Number of attributes.
* @return Array of numeric attributes.
*/
private static DoubleFunction[] initAttributes(int size)
{
DoubleFunction[] nf = new DoubleFunction[size];
for (int i = 0; i < nf.length; i++) nf[i] = new AttributeValue(i);
return nf;
}
/**
* Returns array of weights to input functions with equal coordinates
* normalised to the length 1.
*
* @param size Number of weights.
* @return Array of weights to input functions.
*/
private static double[] initCoefficients(int size)
{
double[] coefficients = new double[size];
for (int i = 0; i < coefficients.length; i++) coefficients[i] = 1/Math.sqrt(size);
return coefficients;
}
/**
* Trains a perceptron in this way that it discriminates
* one decision class from the rest.
*
* @param tab Training table.
* @param dec Decision to be discriminated.
* @param convergence Convergence coefficient.
* @param per Perceptron with weights for initialisation,
* if null the method create new perceptron.
* @return Trained perceptron.
*/
public static Perceptron discriminateOneFromRest(DoubleDataTable tab, double dec, double convergence, Perceptron per)
{
DoubleFunction[] attributes = null;
double[] coefficients = null;
if (per != null) coefficients = per.m_arrCoefficients;
for (DoubleData dObj : tab.getDataObjects())
{
if (attributes == null)
{
attributes = initAttributes(dObj.attributes().noOfAttr());
if (coefficients == null) coefficients = initCoefficients(dObj.attributes().noOfAttr());
}
double sk = 0;
for (int i = 0; i < coefficients.length; i++) sk += coefficients[i]*dObj.get(i);
double delta = 0;
if (((DoubleDataWithDecision)dObj).getDecision()==dec)
{
if (sk <= 0) delta = 2;
}
else
{
if (sk > 0) delta = -2;
}
for (int i = 0; i < coefficients.length; i++) coefficients[i] += convergence*delta*dObj.get(i);
}
return new Perceptron(attributes, coefficients);
}
/**
* Metoda uczaca perceptron, dyskryminujaca dwie klasy decyzyjne.
*
* @param tab Tablica treningowa.
* @param dec1 Wartosc pierwszej decyzji.
* @param dec2 Wartosc drugiej decyzji.
* @param convergance Wspolczynnik zbieznosci.
* @param per Perceptron dostarczajacy poczatkowe wagi wejsc.
* @return Wyuczony perceptron.
*/
/**
* Trains a perceptron in this way that it discriminates
* one decision class from the rest.
*
* @param tab Training table.
* @param dec1 First decision to be discriminated.
* @param dec2 Second decision to be discriminated.
* @param convergence Convergence coefficient.
* @param per Perceptron with weights for initialisation,
* if null the method create new perceptron.
* @return Trained perceptron.
*/
public static Perceptron discriminatePair(DoubleDataTable tab, int dec1, int dec2, double convergence, Perceptron per)
{
DoubleFunction[] attributes = null;
double[] coefficients = null;
if (per != null) coefficients = per.m_arrCoefficients;
for (DoubleData dObj : tab.getDataObjects())
{
if (attributes == null)
{
attributes = initAttributes(dObj.attributes().noOfAttr());
if (coefficients == null) coefficients = initCoefficients(dObj.attributes().noOfAttr());
}
double sk = 0;
for (int i = 0; i < coefficients.length; i++) sk += coefficients[i]*dObj.get(i);
double delta = 0;
double dec = ((DoubleDataWithDecision)dObj).getDecision();
if (dec==dec1 || dec==dec2)
{
if (sk <= 0) delta = 2;
}
else
{
if (sk > 0) delta = -2;
}
for (int i = 0; i < coefficients.length; i++) coefficients[i] += convergence*delta*dObj.get(i);
}
return new Perceptron(attributes, coefficients);
}
}
