eva2/src/eva2/server/go/individuals/GAIndividualDoubleData.java
Johannes Planatscher 7786dbc219 - moved everything in wsi.* to eva2.tools.*
- fixed bug in thread-parallel AbstractOptimization problem (ArrayList -> Vector)
2009-08-06 12:40:42 +00:00

354 lines
14 KiB
Java
Raw Blame History

package eva2.server.go.individuals;
import java.util.BitSet;
import eva2.server.go.individuals.codings.ga.GAStandardCodingDouble;
import eva2.server.go.individuals.codings.ga.InterfaceGADoubleCoding;
import eva2.server.go.operators.crossover.CrossoverGANPoint;
import eva2.server.go.operators.mutation.InterfaceMutation;
import eva2.server.go.operators.mutation.MutateGAStandard;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import eva2.tools.math.RNG;
/** This individual uses a binary genotype to code for double values
* using two alternative encodings.
* Created by IntelliJ IDEA.
* User: streiche
* Date: 24.03.2003
* Time: 16:00:30
* To change this template use Options | File Templates.
*/
public class GAIndividualDoubleData extends AbstractEAIndividual implements InterfaceGAIndividual, InterfaceDataTypeDouble, java.io.Serializable {
private double[] m_Phenotype;
private double[][] m_Range;
protected BitSet m_Genotype;
protected int m_GenotypeLength;
private int m_Precision = 32;
private InterfaceGADoubleCoding m_DoubleCoding = new GAStandardCodingDouble();
public GAIndividualDoubleData() {
this.m_MutationProbability = 0.1;
this.m_MutationOperator = new MutateGAStandard();
this.m_CrossoverProbability = 0.7;
this.m_CrossoverOperator = new CrossoverGANPoint();
this.m_Range = new double[1][2];
this.m_Range[0][0] = -10;
this.m_Range[0][1] = 10;
this.m_GenotypeLength = this.m_Precision;
this.m_Genotype = new BitSet();
}
public GAIndividualDoubleData(GAIndividualDoubleData individual) {
if (individual.m_Phenotype != null) {
this.m_Phenotype = new double[individual.m_Phenotype.length];
System.arraycopy(individual.m_Phenotype, 0, this.m_Phenotype, 0, this.m_Phenotype.length);
}
this.m_GenotypeLength = individual.m_GenotypeLength;
this.m_Genotype = (BitSet) individual.m_Genotype.clone();
this.m_Range = new double[individual.m_Range.length][2];
for (int i = 0; i < this.m_Range.length; i++) {
this.m_Range[i][0] = individual.m_Range[i][0];
this.m_Range[i][1] = individual.m_Range[i][1];
}
// cloning the members of AbstractEAIndividual
this.m_Age = individual.m_Age;
this.m_CrossoverOperator = individual.m_CrossoverOperator;
this.m_CrossoverProbability = individual.m_CrossoverProbability;
this.m_MutationOperator = (InterfaceMutation)individual.m_MutationOperator.clone();
this.m_MutationProbability = individual.m_MutationProbability;
this.m_SelectionProbability = new double[individual.m_SelectionProbability.length];
for (int i = 0; i < this.m_SelectionProbability.length; i++) {
this.m_SelectionProbability[i] = individual.m_SelectionProbability[i];
}
this.m_Precision = individual.m_Precision;
this.m_DoubleCoding = individual.m_DoubleCoding;
this.m_Fitness = new double[individual.m_Fitness.length];
for (int i = 0; i < this.m_Fitness.length; i++) {
this.m_Fitness[i] = individual.m_Fitness[i];
}
cloneAEAObjects((AbstractEAIndividual) individual);
}
public Object clone() {
return (Object) new GAIndividualDoubleData(this);
}
/** This method checks on equality regarding genotypic equality
* @param individual The individual to compare to.
* @return boolean if equal true else false.
*/
public boolean equalGenotypes(AbstractEAIndividual individual) {
if (individual instanceof GAIndividualDoubleData) {
GAIndividualDoubleData indy = (GAIndividualDoubleData) individual;
//@todo Eigendlich k<>nnte ich noch das Koding vergleichen
if (this.m_GenotypeLength != indy.m_GenotypeLength) return false;
if ((this.m_Genotype == null) || (indy.m_Genotype == null)) return false;
if (!this.m_Genotype.equals(indy.m_Genotype)) return false;
for (int i = 0; i < this.m_Range.length; i++) {
if (this.m_Range[i][0] != indy.m_Range[i][0]) return false;
if (this.m_Range[i][1] != indy.m_Range[i][1]) return false;
}
return true;
} else {
return false;
}
}
/************************************************************************************
* InterfaceDataTypeDouble methods
*/
/** This method allows you to request a certain amount of double data
* @param length The lenght of the double[] that is to be optimized
*/
public void setDoubleDataLength (int length) {
double[] newDesPa = new double[length];
double[][] newRange = new double[length][2];
// copy the old values for the decision parameters and the range
for (int i = 0; ((i < newDesPa.length) && (i < this.m_Range.length)); i++) {
newRange[i][0] = this.m_Range[i][0];
newRange[i][1] = this.m_Range[i][1];
}
// if the new length is bigger than the last value fills the extra elements
for (int i = this.m_Range.length; (i < newDesPa.length); i++) {
newRange[i][0] = this.m_Range[this.m_Range.length-1][0];
newRange[i][1] = this.m_Range[this.m_Range.length-1][1];
}
this.m_Range = newRange;
this.m_GenotypeLength = length * this.m_Precision;
// changed 28.08.03 by request of Spieth
// this.m_DecisionParameters = new double[length];
// this.m_Range = new double[length][2];
// for (int i = 0; i < this.m_Range.length; i++) {
// this.m_Range[i][0] = -10;
// this.m_Range[i][1] = 10;
// }
}
/** This method returns the length of the double data set
* @return The number of bits stored
*/
public int size() {
return this.m_Range.length;
}
/** This method will set the range of the double attributes. If range.length
* does not equal doubledata.length only range[i] will be used to set all
* ranges.
* @param range The new range for the double data.
*/
public void SetDoubleRange(double[][] range) {
if (range.length != this.m_Range.length) {
System.out.println("Warning: Trying to set a range of length " + range.length + " to a vector of length "
+ this.m_Range.length + "!\n Use method setDoubleDataLength first!");
}
for (int i = 0; ((i < this.m_Range.length) && (i < range.length)); i++) {
this.m_Range[i][0] = range[i][0];
this.m_Range[i][1] = range[i][1];
}
}
/** This method will return the range for all double attributes.
* @return The range array.
*/
public double[][] getDoubleRange() {
return this.m_Range;
}
/** This method allows you to read the double data
* @return BitSet representing the double data.
*/
public double[] getDoubleData() {
int[] locus = new int[2];
this.m_Phenotype = new double[this.m_Range.length];
for (int i = 0; i < this.m_Range.length; i++) {
locus[0] = i * this.m_Precision;
locus[1] = this.m_Precision;
this.m_Phenotype[i] = this.m_DoubleCoding.decodeValue(this.m_Genotype, this.m_Range[i], locus, false);
}
return this.m_Phenotype;
}
/** This method allows you to read the double data without
* an update from the genotype
* @return double[] representing the double data.
*/
public double[] getDoubleDataWithoutUpdate() {
return this.m_Phenotype;
}
/** This method allows you to set the phenotype data. To change the genotype data,
* use SetDoubleDataLamarckian.
* @param doubleData The new double data.
*/
public void SetDoublePhenotype(double[] doubleData) {
this.m_Phenotype = doubleData;
}
/** This method allows you to set the double data, this can be used for
* memetic algorithms.
* @param doubleData The new double data.
*/
public void SetDoubleGenotype(double[] doubleData) {
this.SetDoublePhenotype(doubleData);
int[] locus = new int[2];
for (int i = 0; i < doubleData.length; i++) {
locus[0] = i * this.m_Precision;
locus[1] = this.m_Precision;
this.m_DoubleCoding.codeValue(doubleData[i], this.m_Range[i], this.m_Genotype, locus);
}
}
/************************************************************************************
* AbstractEAIndividual methods
*/
/** This method will allow a default initialisation of the individual
* @param opt The optimization problem that is to be solved.
*/
public void init(InterfaceOptimizationProblem opt) {
this.defaultInit();
this.m_MutationOperator.init(this, opt);
this.m_CrossoverOperator.init(this, opt);
}
/** This method will init the individual with a given value for the
* phenotype.
* @param obj The initial value for the phenotype
* @param opt The optimization problem that is to be solved.
*/
public void initByValue(Object obj, InterfaceOptimizationProblem opt) {
if (obj instanceof double[]) {
double[] bs = (double[]) obj;
if (bs.length != this.m_Range.length) System.out.println("Init value and requested length doesn't match!");
this.SetDoubleGenotype(bs);
} else {
this.defaultInit();
System.out.println("Initial value for GAIndividualDoubleData is not double[]!");
}
this.m_MutationOperator.init(this, opt);
this.m_CrossoverOperator.init(this, opt);
}
/** This method will return a string description of the GAIndividal
* noteably the Genotype.
* @return A descriptive string
*/
public String getStringRepresentation() {
String result = "";
result += "GAIndividual coding double: (";
result += "Fitness {";
for (int i = 0; i < this.m_Fitness.length; i++) result += this.m_Fitness[i] + ";";
result += "}/SelProb{";
for (int i = 0; i < this.m_SelectionProbability.length; i++) result += this.m_SelectionProbability[i] + ";";
result += "})\n Value: ";
result += "[";
double[] d = this.getDoubleData();
for (int i = 0; i < d.length; i++) {
result += d[i] + "; ";
}
result += "]\n";
result += "{";
for (int i = 0; i < this.m_GenotypeLength; i++) {
if (this.m_Genotype.get(i)) result += "1";
else result += "0";
}
result += "}";
return result;
}
/************************************************************************************
* InterfaceGAIndividual methods
*/
/** This method allows you to read the binary data
* @return BitSet representing the binary data.
*/
public BitSet getBGenotype() {
return this.m_Genotype;
}
/** This method allows you to set the binary data, this can be used for
* memetic algorithms.
* @param binaryData The new binary data.
*/
public void SetBGenotype(BitSet binaryData) {
this.m_Genotype = binaryData;
}
/** This method allows the user to read the length of the genotype.
* This may be necessary since BitSet.lenght only returns the index
* of the last significat bit.
* @return The length of the genotype.
*/
public int getGenotypeLength() {
return this.m_GenotypeLength;
}
/** This method inits the genotpye of the individual
*/
public void defaultInit() {
for (int i = 0; i < this.m_GenotypeLength; i++) {
if (RNG.flipCoin(0.5)) this.m_Genotype.set(i);
else this.m_Genotype.clear(i);
}
}
/** This method performs a simple one point mutation in the genotype
*/
public void defaultMutate() {
int mutationIndex = RNG.randomInt(0, this.m_GenotypeLength);
if (this.m_Genotype.get(mutationIndex)) this.m_Genotype.clear(mutationIndex);
else this.m_Genotype.set(mutationIndex);
}
/**********************************************************************************************************************
* These are for GUI
*/
/** This method allows the CommonJavaObjectEditorPanel to read the
* name to the current object.
* @return The name.
*/
public String getName() {
return "GA individual";
}
/** This method returns a global info string
* @return description
*/
public String globalInfo() {
return "This is a GA individual suited to optimize double values.";
}
/** This method allows you to set the Coding that is to be used, currently either standard binary
* coding or Gray coding.
* @param coding The used genotype coding method
*/
public void setGACoding(InterfaceGADoubleCoding coding) {
this.m_DoubleCoding = coding;
}
public InterfaceGADoubleCoding getGACoding() {
return this.m_DoubleCoding;
}
public String gADoubleCodingTipText() {
return "Choose the coding to use.";
}
/** This method allows you to set the number of mulitruns that are to be performed,
* necessary for stochastic optimizers to ensure reliable results.
* @param precision The number of multiruns that are to be performed
*/
public void setPrecision(int precision) {
this.m_Precision = precision;
}
public int getPrecision() {
return this.m_Precision;
}
public String precisionTipText() {
return "Gives the number of bits to be used to code a double.";
}
}