halfdan/eva2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Refactored IndividualInterface and MutateXXDefault versions. DE/PSO now work on any InterfaceDataTypeDouble

Browse Source
This commit is contained in:
Marcel Kronfeld 2008-09-03 18:03:49 +00:00
parent 0fdb7edc52
commit 9d9f507d13
31 changed files with 264 additions and 489 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
src/eva2/EvAInfo.java
View File

@ -5,11 +5,17 @@ package eva2;
*
* --- Changelog
* 2.030: Added an EnumEditor to access enums easily through the GUI.
* IPOP-ES and RankMuCMA mutator have been added lately (wow!).
* Cleaned up the IndividualInterface and reduced the usage of InterfaceESIndividual. This
* means that, e.g., that DE and PSO now also work on GAIndividualDoubleData. Because this
* requires much time for transcoding, however, this is not useful by itself. Yet it could be
* interesting for combined individuals composed of two data types.
* Cleaned up MutateXXDefault to a single MutateDefault, too.
* 2.029: Tuned the 2d-graphs which now paints quicker and changes size depending on the
* surrounding plot window. Added a preloader-thread to accelerate the GUI at starting time.
* 2.028: Tuned the Population to sort only when necessary on calls to getBestN... Added StatisticsDummy.
* Slightly tuned SimpleProblemWrapper to call initProblem of simple problems if available.
* 2.027: Renamed SetData and SetDataLamarckian from individual datatype interfaces to SetGenotype and SetPhenotype.
* 2.027: Renamed SetData and SetDataLamarckian from individual data type interfaces to SetGenotype and SetPhenotype.
* Repaired the GenericArrayEditor. Population measures can now be plotted in stats.
* 2.026: Added DiversityTerminator and KnownOptimaTerminator, slightly changed InterfaceTerminator for these
* and InterfaceStatistics to provide termination message to text window.

66
src/eva2/server/go/IndividualInterface.java
View File

@ -9,20 +9,60 @@ package eva2.server.go;
* $Date: 2007-12-04 14:22:52 +0100 (Tue, 04 Dec 2007) $
* $Author: mkron $
*/
/*==========================================================================*
* IMPORTS
*==========================================================================*/
/*==========================================================================*
* INTERFACE DECLARATION
*==========================================================================*/
/**
*
* Minimal interface for an EA individual.
*/
public interface IndividualInterface {
public IndividualInterface getClone();
public double[] getFitness();
public void SetFitness (double[] fit);
public double[] getDoubleArray();
public boolean isDominant(double[] otherFitness);
public boolean isDominant(IndividualInterface other);
/**
* Create a clone of the individual instance.
*
* @return a clone of the individual instance
*/
public IndividualInterface getClone();
/**
* Get the fitness array of the individual which may be null if none has been set.
*
* @return the fitness array of the individual
*/
public double[] getFitness();
/**
* Set the fitness array to the given array.
*
* @param fit new fitness of the individual
*/
public void SetFitness (double[] fit);
/**
* Check whether the instance is dominating the given other individual and return
* true in this case.
*
* @param other a second individual of the same type
* @return true if the instance dominates the other individual, else false
*/
public boolean isDominant(double[] fitness);
/**
* Check whether the instance is dominating the given other individual and return
* true in this case.
* Should behave equally to {@link #isDominant(double[])} if called with the fitness
* of the given individual.
*
* @param other a second individual of the same type
* @return true if the instance dominates the other individual, else false
*/
public boolean isDominant(IndividualInterface other);
/**
* Perform a standard mutation operation on the individual. The exact implementation
* depends on the implemented genotype.
*/
public void defaultMutate();
/**
* Initialize the genotype randomly, usually in a uniform distribution.
*/
public void defaultInit();
}

34
src/eva2/server/go/individuals/AbstractEAIndividual.java
View File

@ -22,9 +22,9 @@ import eva2.tools.EVAERROR;
/** This is the abstract EA individual implementing the most important methods giving
* access to mutation and crossover rates and operators, fitness values and selection
* probabilities. All EA individuals should typically extend this abstract EA individual.
* In that case the EA individuals only implement the genotpye and phenotype interfaces.
* The names of the implementation should be build like this:
* (Genotpye)Individual(Phenotype)
* In that case the EA individuals only implement the genotype and phenotype interfaces.
* The names of the implementation should be built like this:
* (Genotype)Individual(Phenotype)
* Thus a binary individual coding double values is named GAIndividualDoubleData and a
* real-valued individual coding binary values is named ESIndividualBinaryData.
* Created by IntelliJ IDEA.
@ -531,7 +531,7 @@ public abstract class AbstractEAIndividual implements IndividualInterface, java.
}
/** This method will allow you to compare two individuals regarding the dominance.
* Note this is dominance! If the individuals are invariant this method will
* Note this is dominance! If the individuals are not comparable this method will
* return false!
* @param indy The individual to compare to.
* @return True if better false else
@ -583,7 +583,7 @@ public abstract class AbstractEAIndividual implements IndividualInterface, java.
}
/** This method will allow you to compare two individuals regarding the dominance.
* Note this is dominance! If the individuals are invariant this method will
* Note this is dominance! If the individuals are not comparable this method will
* return false!
* @param indy The individual to compare to.
* @return True if better false else
@ -607,7 +607,7 @@ public abstract class AbstractEAIndividual implements IndividualInterface, java.
}
/** This method will allow you to compare two individuals regarding the dominance.
* Note this is dominance! If the individuals are invariant this method will
* Note this is dominance! If the individuals are not comparable this method will
* return false!
* @param indy The individual to compare to.
* @return True if better false else
@ -620,7 +620,7 @@ public abstract class AbstractEAIndividual implements IndividualInterface, java.
}
/** This method will allow you to compare two individuals regarding the dominance.
* Note this is dominance! If the individuals are invariant this method will
* Note this is dominance! If the individuals are not comparable this method will
* return false!
*
* @param indy The individual to compare to.
@ -889,16 +889,16 @@ public abstract class AbstractEAIndividual implements IndividualInterface, java.
return (IndividualInterface)this.clone();
}
/** This method is used to get the basic data type of an individual double[].
* @deprecated Since not all EAIndividuals provide double as basic data type
* the fitness can be is returned as default value.
* @see #getFitness()
* @return double[]
*/
public double[] getDoubleArray() {
if (this instanceof InterfaceDataTypeDouble) return ((InterfaceDataTypeDouble)this).getDoubleData();
else return this.getFitness();
}
// /** This method is used to get the basic data type of an individual double[].
// * @deprecated Since not all EAIndividuals provide double as basic data type
// * the fitness can be is returned as default value.
// * @see #getFitness()
// * @return double[]
// */
// public double[] getDoubleArray() {
// if (this instanceof InterfaceDataTypeDouble) return ((InterfaceDataTypeDouble)this).getDoubleData();
// else return this.getFitness();
// }
public boolean isDominantNotEqual(double[] otherFitness) {
return isDominatingFitnessNotEqual(m_Fitness, otherFitness);

9
src/eva2/server/go/individuals/ESIndividualBinaryData.java
View File

@ -257,18 +257,13 @@ public class ESIndividualBinaryData extends AbstractEAIndividual implements Inte
/** This method performs a simple one element mutation on the double vector
*/
public void defaultMutate() {
int mutationIndex = RNG.randomInt(0, this.m_Genotype.length-1);
this.m_Genotype[mutationIndex] += ((this.m_Range[mutationIndex][1] - this.m_Range[mutationIndex][0])/2)*RNG.gaussianDouble(0.05f);
if (this.m_Genotype[mutationIndex] < this.m_Range[mutationIndex][0]) this.m_Genotype[mutationIndex] = this.m_Range[mutationIndex][0];
if (this.m_Genotype[mutationIndex] > this.m_Range[mutationIndex][1]) this.m_Genotype[mutationIndex] = this.m_Range[mutationIndex][1];
ESIndividualDoubleData.defaultMutate(m_Genotype, m_Range);
}
/** This method initializes the double vector
*/
public void defaultInit() {
for (int i = 0; i < this.m_Genotype.length; i++) {
this.m_Genotype[i] = RNG.randomDouble(this.m_Range[i][0], this.m_Range[i][1]);
}
ESIndividualDoubleData.defaultInit(m_Genotype, m_Range);
}
/**********************************************************************************************************************
* These are for GUI

35
src/eva2/server/go/individuals/ESIndividualDoubleData.java
View File

@ -275,17 +275,40 @@ public class ESIndividualDoubleData extends AbstractEAIndividual implements Inte
/** This method performs a simple one element mutation on the double vector
*/
public void defaultMutate() {
int mutationIndex = RNG.randomInt(0, this.m_Genotype.length-1);
this.m_Genotype[mutationIndex] += ((this.m_Range[mutationIndex][1] - this.m_Range[mutationIndex][0])/2)*RNG.gaussianDouble(0.05f);
if (this.m_Genotype[mutationIndex] < this.m_Range[mutationIndex][0]) this.m_Genotype[mutationIndex] = this.m_Range[mutationIndex][0];
if (this.m_Genotype[mutationIndex] > this.m_Range[mutationIndex][1]) this.m_Genotype[mutationIndex] = this.m_Range[mutationIndex][1];
ESIndividualDoubleData.defaultMutate(this.m_Genotype, this.m_Range);
}
/**
* Helper method for default ES mutation. A single, uniformly chosen double entry
* is mutated with a gaussian value.
* If the range constraint is violated, the value is set on the bound.
*
* @param genotype
* @param range
*/
public static void defaultMutate(double[] genotype, double[][] range) {
int mutationIndex = RNG.randomInt(0, genotype.length-1);
genotype[mutationIndex] += ((range[mutationIndex][1] - range[mutationIndex][0])/2)*RNG.gaussianDouble(0.05f);
if (genotype[mutationIndex] < range[mutationIndex][0]) genotype[mutationIndex] = range[mutationIndex][0];
if (genotype[mutationIndex] > range[mutationIndex][1]) genotype[mutationIndex] = range[mutationIndex][1];
}
/** This method initializes the double vector
*/
public void defaultInit() {
for (int i = 0; i < this.m_Genotype.length; i++) {
this.m_Genotype[i] = RNG.randomDouble(this.m_Range[i][0], this.m_Range[i][1]);
ESIndividualDoubleData.defaultInit(m_Genotype, m_Range);
}
/**
* Helper method for initialization. The genotype is distributed uniformly
* within the given range.
*
* @param genotype
* @param range
*/
public static void defaultInit(double[] genotype, double[][] range) {
for (int i = 0; i < genotype.length; i++) {
genotype[i] = RNG.randomDouble(range[i][0], range[i][1]);
}
}

18
src/eva2/server/go/individuals/ESIndividualPermutationData.java
View File

@ -7,7 +7,8 @@ import eva2.server.go.operators.mutation.MutateESGlobal;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import wsi.ra.math.RNG;
/** This individual uses a real-valued genotype to code for a permutations,
/**
* This individual uses a real-valued genotype to code for a permutations,
* the sorting of the real-valued genotype gives the permutation.
* Created by IntelliJ IDEA.
* User: streiche
@ -316,25 +317,20 @@ public class ESIndividualPermutationData extends AbstractEAIndividual implements
}
/** This method performs a simple one element mutation on the double vector
/** This method performs a one element mutation on every permutation coded by a double vector.
*/
public void defaultMutate() {
for (int i = 0; i < m_Genotype.length; i++) {
int mutationIndex = RNG.randomInt(0, this.m_Genotype[i].length-1);
this.m_Genotype[i][mutationIndex] += ((this.m_Range[i][mutationIndex][1] - this.m_Range[i][mutationIndex][0])/2)*RNG.gaussianDouble(0.05f);
if (this.m_Genotype[i][mutationIndex] < this.m_Range[i][mutationIndex][0]) this.m_Genotype[i][mutationIndex] = this.m_Range[i][mutationIndex][0];
if (this.m_Genotype[i][mutationIndex] > this.m_Range[i][mutationIndex][1]) this.m_Genotype[i][mutationIndex] = this.m_Range[i][mutationIndex][1];
ESIndividualDoubleData.defaultMutate(m_Genotype[i], m_Range[i]);
}
}
/** This method initializes the double vector
/**
* This method initializes the double vector
*/
public void defaultInit() {
for (int i = 0; i < this.m_Genotype.length; i++) {
for (int j = 0; j < this.m_Genotype[i].length; j++) {
this.m_Genotype[i][j] = RNG.randomDouble(this.m_Range[i][j][0], this.m_Range[i][j][1]);
}
ESIndividualDoubleData.defaultInit(m_Genotype[i], m_Range[i]);
}
}

10
src/eva2/server/go/individuals/GAESIndividualBinaryDoubleData.java
View File

@ -75,6 +75,11 @@ public class GAESIndividualBinaryDoubleData extends AbstractEAIndividual impleme
((AbstractEAIndividual)this.m_BitSet).init(opt);
}
public void defaultInit() {
((AbstractEAIndividual)this.m_Numbers).defaultInit();
((AbstractEAIndividual)this.m_BitSet).defaultInit();
}
/** This method will init the individual with a given value for the
* phenotype.
* @param obj The initial value for the phenotype
@ -103,6 +108,11 @@ public class GAESIndividualBinaryDoubleData extends AbstractEAIndividual impleme
if (RNG.flipCoin(this.m_MutationProbability))((AbstractEAIndividual)this.m_BitSet).mutate();
}
public void defaultMutate() {
((AbstractEAIndividual)this.m_Numbers).defaultMutate();
((AbstractEAIndividual)this.m_BitSet).defaultMutate();
}
/** This method will mate the Individual with given other individuals
* of the same type.
* @param partners The possible partners

13
src/eva2/server/go/individuals/GAIndividualBinaryData.java
View File

@ -3,16 +3,11 @@ package eva2.server.go.individuals;
import java.util.BitSet;
import eva2.server.go.operators.crossover.CrossoverGADefault;
import eva2.server.go.operators.crossover.CrossoverGANPoint;
import eva2.server.go.operators.crossover.InterfaceCrossover;
import eva2.server.go.operators.crossover.NoCrossover;
import eva2.server.go.operators.mutation.InterfaceMutation;
import eva2.server.go.operators.mutation.MutateGADefault;
import eva2.server.go.operators.mutation.MutateGAStandard;
import eva2.server.go.operators.mutation.NoMutation;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import wsi.ra.math.RNG;
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;
/** This individual uses a binary genotype to code for binary values.
* Created by IntelliJ IDEA.

5
src/eva2/server/go/individuals/GAIndividualDoubleData.java
View File

@ -3,16 +3,13 @@ package eva2.server.go.individuals;
import java.util.BitSet;
import wsi.ra.math.RNG;
import eva2.server.go.individuals.codings.ga.GAStandardCodingDouble;
import eva2.server.go.individuals.codings.ga.InterfaceGADoubleCoding;
import eva2.server.go.operators.crossover.CrossoverGADefault;
import eva2.server.go.operators.crossover.CrossoverGANPoint;
import eva2.server.go.operators.crossover.InterfaceCrossover;
import eva2.server.go.operators.mutation.InterfaceMutation;
import eva2.server.go.operators.mutation.MutateGADefault;
import eva2.server.go.operators.mutation.MutateGAStandard;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import wsi.ra.math.RNG;
/** This individual uses a binary genotype to code for double values
* using two alternative encodings.

23
src/eva2/server/go/individuals/GAPIndividualProgramData.java
View File

@ -1,20 +1,11 @@
package eva2.server.go.individuals;
import java.util.ArrayList;
import eva2.server.go.individuals.codings.gp.AbstractGPNode;
import eva2.server.go.individuals.codings.gp.GPArea;
import wsi.ra.math.RNG;
import eva2.server.go.individuals.codings.gp.InterfaceProgram;
import eva2.server.go.operators.crossover.CrossoverESDefault;
import eva2.server.go.operators.crossover.CrossoverGPDefault;
import eva2.server.go.operators.crossover.InterfaceCrossover;
import eva2.server.go.operators.mutation.InterfaceMutation;
import eva2.server.go.operators.mutation.MutateESDefault;
import eva2.server.go.operators.mutation.MutateGPDefault;
import eva2.server.go.populations.Population;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import wsi.ra.math.RNG;
/** This individual combines a real-valued phenotype with a tree-based phenotype.
* Created by IntelliJ IDEA.
@ -82,7 +73,12 @@ public class GAPIndividualProgramData extends AbstractEAIndividual implements In
((AbstractEAIndividual)this.m_Numbers).init(opt);
((AbstractEAIndividual)this.m_Program).init(opt);
}
public void defaultInit() {
((AbstractEAIndividual)this.m_Numbers).defaultInit();
((AbstractEAIndividual)this.m_Program).defaultInit();
}
/** This method will init the individual with a given value for the
* phenotype.
* @param obj The initial value for the phenotype
@ -111,6 +107,11 @@ public class GAPIndividualProgramData extends AbstractEAIndividual implements In
if (RNG.flipCoin(this.m_MutationProbability))((AbstractEAIndividual)this.m_Program).mutate();
}
public void defaultMutate() {
((AbstractEAIndividual)this.m_Numbers).defaultMutate();
((AbstractEAIndividual)this.m_Program).defaultMutate();
}
/** This method will mate the Individual with given other individuals
* of the same type.
* @param partners The possible partners

8
src/eva2/server/go/individuals/GEIndividualProgramData.java
View File

@ -1,17 +1,17 @@
package eva2.server.go.individuals;
import java.util.BitSet;
import java.util.ArrayList;
import java.util.BitSet;
import wsi.ra.math.RNG;
import eva2.server.go.individuals.codings.gp.AbstractGPNode;
import eva2.server.go.individuals.codings.gp.GPArea;
import eva2.server.go.individuals.codings.gp.InterfaceProgram;
import eva2.server.go.operators.crossover.CrossoverGADefault;
import eva2.server.go.operators.mutation.InterfaceMutation;
import eva2.server.go.operators.mutation.MutateGADefault;
import eva2.server.go.operators.mutation.MutateDefault;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import wsi.ra.math.RNG;
/** This individual uses a binary genotype to code for a tree-based representation
* using a BNF grammar, see also Grammatical Evolution.
@ -42,7 +42,7 @@ public class GEIndividualProgramData extends AbstractEAIndividual implements Int
this.m_Area = new GPArea[1];
this.m_GenotypeLengthPerProgram = 240;
this.m_Genotype = new BitSet();
this.m_MutationOperator = new MutateGADefault();
this.m_MutationOperator = new MutateDefault();
this.m_CrossoverOperator = new CrossoverGADefault();
this.m_MutationProbability = 0.5;
this.m_CrossoverProbability = 0.5;

6
src/eva2/server/go/individuals/GIIndividualIntegerData.java
View File

@ -1,10 +1,10 @@
package eva2.server.go.individuals;
import wsi.ra.math.RNG;
import eva2.server.go.operators.crossover.CrossoverGIDefault;
import eva2.server.go.operators.mutation.InterfaceMutation;
import eva2.server.go.operators.mutation.MutateGIDefault;
import eva2.server.go.operators.mutation.MutateDefault;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import wsi.ra.math.RNG;
/** This individual uses a integer genotype to code for integer values.
* Created by IntelliJ IDEA.
@ -21,7 +21,7 @@ public class GIIndividualIntegerData extends AbstractEAIndividual implements Int
public GIIndividualIntegerData() {
this.m_MutationProbability = 0.1;
this.m_MutationOperator = new MutateGIDefault();
this.m_MutationOperator = new MutateDefault();
this.m_CrossoverProbability = 0.7;
this.m_CrossoverOperator = new CrossoverGIDefault();
this.m_Range = new int[10][2];

10
src/eva2/server/go/individuals/GIOBGAIndividualIntegerPermutationData.java
View File

@ -74,6 +74,11 @@ public class GIOBGAIndividualIntegerPermutationData extends AbstractEAIndividual
((AbstractEAIndividual)this.m_Integer).init(opt);
((AbstractEAIndividual)this.m_Permutation).init(opt);
}
public void defaultInit() {
((AbstractEAIndividual)this.m_Integer).defaultInit();
((AbstractEAIndividual)this.m_Permutation).defaultInit();
}
/** This method will init the individual with a given value for the
* phenotype.
@ -103,6 +108,11 @@ public class GIOBGAIndividualIntegerPermutationData extends AbstractEAIndividual
if (RNG.flipCoin(this.m_MutationProbability))((AbstractEAIndividual)this.m_Permutation).mutate();
}
public void defaultMutate() {
((AbstractEAIndividual)this.m_Integer).defaultMutate();
((AbstractEAIndividual)this.m_Permutation).defaultMutate();
}
/** This method will mate the Individual with given other individuals
* of the same type.
* @param partners The possible partners

7
src/eva2/server/go/individuals/GPIndividualProgramData.java
View File

@ -3,16 +3,15 @@ package eva2.server.go.individuals;
import java.util.ArrayList;
import wsi.ra.math.RNG;
import eva2.server.go.individuals.codings.gp.AbstractGPNode;
import eva2.server.go.individuals.codings.gp.GPArea;
import eva2.server.go.individuals.codings.gp.InterfaceProgram;
import eva2.server.go.operators.crossover.CrossoverGPDefault;
import eva2.server.go.operators.mutation.InterfaceMutation;
import eva2.server.go.operators.mutation.MutateGPDefault;
import eva2.server.go.operators.mutation.MutateDefault;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import wsi.ra.math.RNG;
import eva2.tools.EVAERROR;
import eva2.tools.EVAHELP;
/** This individual uses a tree-based genotype to code for program trees.
* Created by IntelliJ IDEA.
@ -35,7 +34,7 @@ public class GPIndividualProgramData extends AbstractEAIndividual implements Int
this.m_Area = new GPArea[1];
m_Area[0] = new GPArea();
this.m_Genotype = new AbstractGPNode[1];
this.m_MutationOperator = new MutateGPDefault();
this.m_MutationOperator = new MutateDefault();
this.m_CrossoverOperator = new CrossoverGPDefault();
}

7
src/eva2/server/go/individuals/InterfaceESIndividual.java
View File

@ -33,11 +33,4 @@ public interface InterfaceESIndividual {
*/
public double[][] getDoubleRange();
/** This method performs a simple one element mutation on the double vector
*/
public void defaultMutate();
/** This method initializes the double vector
*/
public void defaultInit();
}

9
src/eva2/server/go/individuals/InterfaceGAIndividual.java
View File

@ -11,7 +11,7 @@ import java.util.BitSet;
* Time: 14:25:24
* To change this template use Options | File Templates.
*/
public interface InterfaceGAIndividual {
public interface InterfaceGAIndividual {
/** This method will allow the user to read the GA genotype
* @return BitSet
@ -33,11 +33,4 @@ public interface InterfaceGAIndividual {
*/
public int getGenotypeLength();
/** This method performs a simple one point mutation in the genotype
*/
public void defaultMutate();
/** This method initializes the GA genotype randomly
*/
public void defaultInit();
}

9
src/eva2/server/go/individuals/InterfaceGIIndividual.java
View File

@ -1,6 +1,5 @@
package eva2.server.go.individuals;
import java.util.BitSet;
/** This interface gives access to a integer genotype and should
* only be used by mutation and crossover operators.
@ -48,12 +47,4 @@ public interface InterfaceGIIndividual {
* @return The length of the genotype.
*/
public int getGenotypeLength();
/** This method performs a simple one point mutation in the genotype
*/
public void defaultMutate();
/** This method initializes the GA genotype randomly
*/
public void defaultInit();
}

7
src/eva2/server/go/individuals/InterfaceGPIndividual.java
View File

@ -33,11 +33,4 @@ public interface InterfaceGPIndividual {
*/
public Object[] getFunctionArea();
/** This method performs a simple one element mutation on the program
*/
public void defaultMutate();
/** This method initializes the program
*/
public void defaultInit();
}

10
src/eva2/server/go/individuals/InterfaceOBGAIndividual.java
View File

@ -27,14 +27,4 @@ public interface InterfaceOBGAIndividual {
* @param b int[] new genotype
*/
public void SetOBGenotype(int[][] b);
/**
* defaultMutate perfoms a mutation by flipping two elements in the permutation
*/
public void defaultMutate();
/** This method initializes the program
*/
public void defaultInit();
}

6
src/eva2/server/go/individuals/codings/ga/InterfaceGADoubleCoding.java
View File

@ -12,7 +12,7 @@ import java.util.BitSet;
public interface InterfaceGADoubleCoding {
/** This method decodes a part of a given BitSet into a double value. This method may change the contens
* of the BitSet if it doesn't describe a valid value.
* The method checks wether or not the value is within the given range.
* The method checks whether or not the value is within the given range.
* @param refBitSet The BitSet where the questioned value is stored.
* @param range The allowed range of the value.
* @param locus The position and length on the BitSet that is to be decoded.
@ -21,9 +21,9 @@ public interface InterfaceGADoubleCoding {
*/
public double decodeValue(BitSet refBitSet, double[] range, int[] locus, boolean correction);
/** This method codes a given int value directly into a BitSet at
/** This method codes a given double value directly into a BitSet at
* the position which is specified by locus.
* The method checks wether or not the value is within the given range.
* The method checks whether or not the value is within the given range.
* @param value The value to be coded.
* @param range The allowed range of the value.
* @param refBitSet The BitSet where the questioned value is stored.

41
src/eva2/server/go/operators/mutation/MutateGIDefault.java → src/eva2/server/go/operators/mutation/MutateDefault.java
View File

@ -1,24 +1,23 @@
package eva2.server.go.operators.mutation;
import eva2.server.go.IndividualInterface;
import eva2.server.go.individuals.AbstractEAIndividual;
import eva2.server.go.individuals.InterfaceGIIndividual;
import eva2.server.go.individuals.InterfaceESIndividual;
import eva2.server.go.populations.Population;
import eva2.server.go.problems.InterfaceOptimizationProblem;
/**
* Created by IntelliJ IDEA.
* User: streiche
* Date: 18.05.2005
* Time: 17:08:49
* To change this template use File | Settings | File Templates.
* Mutate individuals using the default operator implemented by the individuals themselves.
*
*/
public class MutateGIDefault implements InterfaceMutation, java.io.Serializable {
public class MutateDefault implements InterfaceMutation, java.io.Serializable {
/** This method will enable you to clone a given mutation operator
* @return The clone
*/
public Object clone() {
return new MutateGIDefault();
return new MutateDefault();
}
/** This method allows you to evaluate wether two mutation operators
@ -26,7 +25,7 @@ public class MutateGIDefault implements InterfaceMutation, java.io.Serializable
* @param mutator The other mutation operator
*/
public boolean equals(Object mutator) {
if (mutator instanceof MutateGIDefault) return true;
if (mutator instanceof MutateDefault) return true;
else return false;
}
@ -38,6 +37,14 @@ public class MutateGIDefault implements InterfaceMutation, java.io.Serializable
}
/** This method will mutate a given AbstractEAIndividual. If the individual
* doesn't implement InterfaceGAIndividual nothing happens.
* @param individual The individual that is to be mutated
*/
public void mutate(AbstractEAIndividual individual) {
if (individual instanceof IndividualInterface) ((IndividualInterface)individual).defaultMutate();
}
/** This method allows you to perform either crossover on the strategy parameters
* or to deal in some other way with the crossover event.
* @param indy1 The original mother
@ -47,20 +54,12 @@ public class MutateGIDefault implements InterfaceMutation, java.io.Serializable
// nothing to do here
}
/** This method will mutate a given AbstractEAIndividual. If the individual
* doesn't implement InterfaceGAIndividual nothing happens.
* @param individual The individual that is to be mutated
*/
public void mutate(AbstractEAIndividual individual) {
if (individual instanceof InterfaceGIIndividual) ((InterfaceGIIndividual)individual).defaultMutate();
}
/** This method allows you to get a string representation of the mutation
* operator
* @return A descriptive string.
*/
public String getStringRepresentation() {
return "GI default mutation";
return "Default mutation";
}
/**********************************************************************************************************************
@ -71,12 +70,12 @@ public class MutateGIDefault implements InterfaceMutation, java.io.Serializable
* @return The name.
*/
public String getName() {
return "GI default mutation";
return "Default mutation";
}
/** This method returns a global info string
* @return description
*/
public String globalInfo() {
return "The default mutation alters one element of the int attributes.";
return "The default mutation just uses the default method implemented in the individual.";
}
}
}

85
src/eva2/server/go/operators/mutation/MutateESDefault.java
View File

@ -1,85 +0,0 @@
package eva2.server.go.operators.mutation;
import eva2.server.go.individuals.AbstractEAIndividual;
import eva2.server.go.individuals.InterfaceESIndividual;
import eva2.server.go.populations.Population;
import eva2.server.go.problems.InterfaceOptimizationProblem;
/**
* Created by IntelliJ IDEA.
* User: streiche
* Date: 25.03.2003
* Time: 10:58:44
* To change this template use Options | File Templates.
*/
public class MutateESDefault implements InterfaceMutation, java.io.Serializable {
/** This method will enable you to clone a given mutation operator
* @return The clone
*/
public Object clone() {
return new MutateESDefault();
}
/** This method allows you to evaluate wether two mutation operators
* are actually the same.
* @param mutator The other mutation operator
*/
public boolean equals(Object mutator) {
if (mutator instanceof MutateESDefault) return true;
else return false;
}
/** This method allows you to init the mutation operator
* @param individual The individual that will be mutated.
* @param opt The optimization problem.
*/
public void init(AbstractEAIndividual individual, InterfaceOptimizationProblem opt){
}
/** This method will mutate a given AbstractEAIndividual. If the individual
* doesn't implement InterfaceGAIndividual nothing happens.
* @param individual The individual that is to be mutated
*/
public void mutate(AbstractEAIndividual individual) {
//System.out.println("Before Mutate: " +((GAIndividual)individual).getSolutionRepresentationFor());
if (individual instanceof InterfaceESIndividual) ((InterfaceESIndividual)individual).defaultMutate();
//System.out.println("After Mutate: " +((GAIndividual)individual).getSolutionRepresentationFor());
}
/** This method allows you to perform either crossover on the strategy parameters
* or to deal in some other way with the crossover event.
* @param indy1 The original mother
* @param partners The original partners
*/
public void crossoverOnStrategyParameters(AbstractEAIndividual indy1, Population partners) {
// nothing to do here
}
/** This method allows you to get a string representation of the mutation
* operator
* @return A descriptive string.
*/
public String getStringRepresentation() {
return "ES default mutation";
}
/**********************************************************************************************************************
* These are for GUI
*/
/** This method allows the CommonJavaObjectEditorPanel to read the
* name to the current object.
* @return The name.
*/
public String getName() {
return "ES default mutation";
}
/** This method returns a global info string
* @return description
*/
public String globalInfo() {
return "The default mutation just uses the default method implemented in the individual.";
}
}

88
src/eva2/server/go/operators/mutation/MutateGADefault.java
View File

@ -1,88 +0,0 @@
package eva2.server.go.operators.mutation;
import eva2.server.go.individuals.AbstractEAIndividual;
import eva2.server.go.individuals.InterfaceGAIndividual;
import eva2.server.go.populations.Population;
import eva2.server.go.problems.InterfaceOptimizationProblem;
/**
* This operator performs one-point mutation.
* Created by IntelliJ IDEA.
*
* User: streiche
* Date: 18.03.2003
* Time: 12:36:08
* To change this template use Options | File Templates.
*/
public class MutateGADefault implements InterfaceMutation, java.io.Serializable {
/** This method will enable you to clone a given mutation operator
* @return The clone
*/
public Object clone() {
return new MutateGADefault();
}
/** This method allows you to evaluate wether two mutation operators
* are actually the same.
* @param mutator The other mutation operator
*/
public boolean equals(Object mutator) {
if (mutator instanceof MutateGADefault) return true;
else return false;
}
/** This method allows you to init the mutation operator
* @param individual The individual that will be mutated.
* @param opt The optimization problem.
*/
public void init(AbstractEAIndividual individual, InterfaceOptimizationProblem opt) {
}
/** This method will mutate a given AbstractEAIndividual. If the individual
* doesn't implement InterfaceGAIndividual nothing happens.
* @param individual The individual that is to be mutated
*/
public void mutate(AbstractEAIndividual individual) {
//System.out.println("Before Mutate: " +((GAIndividual)individual).getSolutionRepresentationFor());
if (individual instanceof InterfaceGAIndividual) ((InterfaceGAIndividual)individual).defaultMutate();
//System.out.println("After Mutate: " +((GAIndividual)individual).getSolutionRepresentationFor());
}
/** This method allows you to perform either crossover on the strategy parameters
* or to deal in some other way with the crossover event.
* @param indy1 The original mother
* @param partners The original partners
*/
public void crossoverOnStrategyParameters(AbstractEAIndividual indy1, Population partners) {
// nothing to do here
}
/** This method allows you to get a string representation of the mutation
* operator
* @return A descriptive string.
*/
public String getStringRepresentation() {
return "GA default mutation";
}
/**********************************************************************************************************************
* 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 default mutation";
}
/** This method returns a global info string
* @return description
*/
public String globalInfo() {
return "The default mutation switches one bit of the GA genotype.";
}
}

8
src/eva2/server/go/operators/mutation/MutateGPAdaptive.java
View File

@ -1,14 +1,12 @@
package eva2.server.go.operators.mutation;
import java.util.BitSet;
import wsi.ra.math.RNG;
import eva2.server.go.IndividualInterface;
import eva2.server.go.individuals.AbstractEAIndividual;
import eva2.server.go.individuals.InterfaceGAIndividual;
import eva2.server.go.individuals.InterfaceGPIndividual;
import eva2.server.go.populations.Population;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import wsi.ra.math.RNG;
/**
* Created by IntelliJ IDEA.
@ -73,7 +71,7 @@ public class MutateGPAdaptive implements InterfaceMutation, java.io.Serializable
this.m_MutationStep = this.m_MutationStep * Math.exp(this.m_Tau1 * RNG.gaussianDouble(1) + this.m_Tau2 * RNG.gaussianDouble(1));
if (this.m_MutationStep < this.m_LowerLimitStepSize) this.m_MutationStep = this.m_LowerLimitStepSize;
if (this.m_MutationStep > 1) this.m_MutationStep = 1;
if (RNG.flipCoin(this.m_MutationStep)) ((InterfaceGPIndividual)individual).defaultMutate();
if (RNG.flipCoin(this.m_MutationStep)) ((IndividualInterface)individual).defaultMutate();
}
//System.out.println("After Mutate: " +((GAIndividual)individual).getSolutionRepresentationFor());
}

89
src/eva2/server/go/operators/mutation/MutateGPDefault.java
View File

@ -1,89 +0,0 @@
package eva2.server.go.operators.mutation;
import eva2.server.go.individuals.AbstractEAIndividual;
import eva2.server.go.individuals.InterfaceGPIndividual;
import eva2.server.go.populations.Population;
import eva2.server.go.problems.InterfaceOptimizationProblem;
/**
* Created by IntelliJ IDEA.
* User: streiche
* Date: 04.04.2003
* Time: 16:38:31
* To change this template use Options | File Templates.
*/
public class MutateGPDefault implements InterfaceMutation, java.io.Serializable {
public MutateGPDefault() {
}
/** This method will enable you to clone a given mutation operator
* @return The clone
*/
public Object clone() {
return new MutateGPDefault();
}
/** This method allows you to evaluate wether two mutation operators
* are actually the same.
* @param mutator The other mutation operator
*/
public boolean equals(Object mutator) {
if (mutator instanceof MutateGPDefault) return true;
else return false;
}
/** This method allows you to init the mutation operator
* @param individual The individual that will be mutated.
* @param opt The optimization problem.
*/
public void init(AbstractEAIndividual individual, InterfaceOptimizationProblem opt) {
}
/** This method will mutate a given AbstractEAIndividual. If the individual
* doesn't implement InterfaceGAIndividual nothing happens.
* @param individual The individual that is to be mutated
*/
public void mutate(AbstractEAIndividual individual) {
// System.out.println("Before Mutate: " +((InterfaceGPIndividual)individual).getPGenotype()[0].getStringRepresentation());
// System.out.println("Length: " +((InterfaceGPIndividual)individual).getPGenotype()[0].getNumberOfNodes());
if (individual instanceof InterfaceGPIndividual) ((InterfaceGPIndividual)individual).defaultMutate();
// System.out.println("After Mutate: " +((InterfaceGPIndividual)individual).getPGenotype()[0].getStringRepresentation());
// System.out.println("Length: " +((InterfaceGPIndividual)individual).getPGenotype()[0].getNumberOfNodes());
}
/** This method allows you to perform either crossover on the strategy parameters
* or to deal in some other way with the crossover event.
* @param indy1 The original mother
* @param partners The original partners
*/
public void crossoverOnStrategyParameters(AbstractEAIndividual indy1, Population partners) {
// nothing to do here
}
/** This method allows you to get a string representation of the mutation
* operator
* @return A descriptive string.
*/
public String getStringRepresentation() {
return "GP default mutation";
}
/**********************************************************************************************************************
* These are for GUI
*/
/** This method allows the CommonJavaObjectEditorPanel to read the
* name to the current object.
* @return The name.
*/
public String getName() {
return "GP default mutation";
}
/** This method returns a global info string
* @return description
*/
public String globalInfo() {
return "The default mutation replaces a random subtree using init, also called headless chicken mutation.";
}
}

28
src/eva2/server/go/strategies/DifferentialEvolution.java
View File

@ -6,7 +6,7 @@ import wsi.ra.math.RNG;
import eva2.gui.GenericObjectEditor;
import eva2.server.go.InterfacePopulationChangedEventListener;
import eva2.server.go.individuals.AbstractEAIndividual;
import eva2.server.go.individuals.InterfaceESIndividual;
import eva2.server.go.individuals.InterfaceDataTypeDouble;
import eva2.server.go.operators.selection.replacement.ReplacementCrowding;
import eva2.server.go.populations.InterfaceSolutionSet;
import eva2.server.go.populations.Population;
@ -166,11 +166,11 @@ public class DifferentialEvolution implements InterfaceOptimizer, java.io.Serial
* @param indy The current individual
* @return the delta vector
*/
private double[] fetchDeltaBest(Population pop, InterfaceESIndividual indy) {
private double[] fetchDeltaBest(Population pop, InterfaceDataTypeDouble indy) {
double[] x1, result;
AbstractEAIndividual xbIndy;
x1 = indy.getDGenotype();
x1 = indy.getDoubleData();
result = new double[x1.length];
if (m_Problem instanceof AbstractMultiObjectiveOptimizationProblem) {
// implements MODE for the multi-objective case: a dominating individual is selected for difference building
@ -220,21 +220,21 @@ public class DifferentialEvolution implements InterfaceOptimizer, java.io.Serial
public AbstractEAIndividual generateNewIndividual(Population pop) {
int firstParentIndex;
AbstractEAIndividual indy;
InterfaceESIndividual esIndy;
InterfaceDataTypeDouble esIndy;
if (doLogParents) parents = new Vector<AbstractEAIndividual>();
else parents = null;
try {
// select one random indy as starting individual. its a parent in any case.
firstParentIndex = RNG.randomInt(0, pop.size()-1);
indy = (AbstractEAIndividual)(pop.getEAIndividual(firstParentIndex)).getClone();
esIndy = (InterfaceESIndividual)indy;
esIndy = (InterfaceDataTypeDouble)indy;
} catch (java.lang.ClassCastException e) {
System.err.println("Differential Evolution currently requires InterfaceESIndividual as basic data type!");
return (AbstractEAIndividual)((AbstractEAIndividual)pop.get(RNG.randomInt(0, pop.size()-1))).getClone();
}
double[] nX, vX, oX;
oX = esIndy.getDGenotype();
vX = esIndy.getDGenotype();
oX = esIndy.getDoubleData();
vX = esIndy.getDoubleData();
nX = new double[oX.length];
switch (this.m_DEType.getSelectedTag().getID()) {
case 0 : {
@ -274,11 +274,11 @@ public class DifferentialEvolution implements InterfaceOptimizer, java.io.Serial
if (RNG.flipCoin(this.m_Mt)) {
double[] xk, xl;
double p, pj, pk, pl;
InterfaceESIndividual indy1 = null, indy2 = null;
InterfaceDataTypeDouble indy1 = null, indy2 = null;
try {
// and i got indy!
indy1 = (InterfaceESIndividual)pop.get(RNG.randomInt(0, pop.size()-1));
indy2 = (InterfaceESIndividual)pop.get(RNG.randomInt(0, pop.size()-1));
indy1 = (InterfaceDataTypeDouble)pop.get(RNG.randomInt(0, pop.size()-1));
indy2 = (InterfaceDataTypeDouble)pop.get(RNG.randomInt(0, pop.size()-1));
if (parents != null) {
parents.add((AbstractEAIndividual)indy1);
parents.add((AbstractEAIndividual)indy2);
@ -286,8 +286,8 @@ public class DifferentialEvolution implements InterfaceOptimizer, java.io.Serial
} catch (java.lang.ClassCastException e) {
EVAERROR.errorMsgOnce("Differential Evolution currently requires InterfaceESIndividual as basic data type!");
}
xk = indy1.getDGenotype();
xl = indy2.getDGenotype();
xk = indy1.getDoubleData();
xl = indy2.getDoubleData();
p = Math.abs(((AbstractEAIndividual)esIndy).getFitness(0)) + Math.abs(((AbstractEAIndividual)indy1).getFitness(0)) + Math.abs(((AbstractEAIndividual)indy2).getFitness(0));
pj = Math.abs(((AbstractEAIndividual)esIndy).getFitness(0))/p;
pk = Math.abs(((AbstractEAIndividual)indy1).getFitness(0))/p;
@ -316,7 +316,7 @@ public class DifferentialEvolution implements InterfaceOptimizer, java.io.Serial
}
}
// setting the new genotype and fitness
esIndy.SetDGenotype(nX);
esIndy.SetDoubleGenotype(nX);
indy.SetAge(0);
double[] fit = new double[1];
fit[0] = 0;
@ -339,7 +339,7 @@ public class DifferentialEvolution implements InterfaceOptimizer, java.io.Serial
private double[] getGenotype(AbstractEAIndividual indy) {
try {
return ((InterfaceESIndividual)indy).getDGenotype();
return ((InterfaceDataTypeDouble)indy).getDoubleData();
} catch (java.lang.ClassCastException e) {
EVAERROR.errorMsgOnce("Differential Evolution currently requires InterfaceESIndividual as basic data type!");
return null;

10
src/eva2/server/go/strategies/DynamicParticleSwarmOptimization.java
View File

@ -132,11 +132,11 @@ public class DynamicParticleSwarmOptimization extends ParticleSwarmOptimization
* @param best The best individual found so far.
*/
private void updateQuantumIndividual(int index, AbstractEAIndividual indy, Population pop) {
InterfaceESIndividual endy = (InterfaceESIndividual) indy;
InterfaceDataTypeDouble endy = (InterfaceDataTypeDouble) indy;
// search for the local best position
double[] localBestPosition;
double[] position = endy.getDGenotype();
double[] position = endy.getDoubleData();
localBestPosition = findNeighbourhoodOptimum(index, pop);
@ -157,7 +157,7 @@ public class DynamicParticleSwarmOptimization extends ParticleSwarmOptimization
}
if (indy instanceof InterfaceDataTypeDouble) ((InterfaceDataTypeDouble)indy).SetDoubleGenotype(newPos);
else endy.SetDGenotype(newPos);
else endy.SetDoubleGenotype(newPos);
resetFitness(indy);
@ -392,7 +392,7 @@ public class DynamicParticleSwarmOptimization extends ParticleSwarmOptimization
*/
protected void updateIndividual(int index, AbstractEAIndividual indy, Population pop) {
if (index != detectAnchor) { // only for non anchor individuals (its -1 if other detect strategy is used)
if (indy instanceof InterfaceESIndividual) {
if (indy instanceof InterfaceDataTypeDouble) {
int type=(Integer)indy.getData(partTypeKey);
switch (type) {
case quantumType:
@ -424,7 +424,7 @@ public class DynamicParticleSwarmOptimization extends ParticleSwarmOptimization
// }
if (doSpeedAdaptation) {
adaptTrackingSpeed(((InterfaceESIndividual)population.get(0)).getDoubleRange());
adaptTrackingSpeed(((InterfaceDataTypeDouble)population.get(0)).getDoubleRange());
}
// if (m_Problem instanceof DynLocalizationProblem) {
// ((DynLocalizationProblem)m_Problem).adaptPSOByPopulation(population, this);

4
src/eva2/server/go/strategies/ParticleFilterOptimization.java
View File

@ -134,8 +134,8 @@ public class ParticleFilterOptimization implements InterfaceOptimizer, java.io.S
indy.SetFitness(0, 0);
if (this.withShow) {
InterfaceESIndividual endy = (InterfaceESIndividual) indy;
double[] curPosition = endy.getDGenotype();
InterfaceDataTypeDouble endy = (InterfaceDataTypeDouble) indy;
double[] curPosition = endy.getDoubleData();
myPlot.setUnconnectedPoint(curPosition[0], curPosition[1], indCount);
myPlot.setConnectedPoint(curPosition[0], curPosition[1], indCount);

52
src/eva2/server/go/strategies/ParticleSwarmOptimization.java
View File

@ -3,6 +3,10 @@ package eva2.server.go.strategies;
import java.util.Arrays;
import java.util.Vector;
import wsi.ra.chart2d.DPoint;
import wsi.ra.chart2d.DPointSet;
import wsi.ra.math.RNG;
import wsi.ra.math.Jama.Matrix;
import eva2.gui.BeanInspector;
import eva2.gui.GenericObjectEditor;
import eva2.gui.TopoPlot;
@ -10,7 +14,6 @@ import eva2.server.go.InterfacePopulationChangedEventListener;
import eva2.server.go.individuals.AbstractEAIndividual;
import eva2.server.go.individuals.AbstractEAIndividualComparator;
import eva2.server.go.individuals.InterfaceDataTypeDouble;
import eva2.server.go.individuals.InterfaceESIndividual;
import eva2.server.go.operators.distancemetric.PhenotypeMetric;
import eva2.server.go.populations.InterfaceSolutionSet;
import eva2.server.go.populations.Population;
@ -18,15 +21,10 @@ import eva2.server.go.populations.SolutionSet;
import eva2.server.go.problems.F1Problem;
import eva2.server.go.problems.Interface2DBorderProblem;
import eva2.server.go.problems.InterfaceOptimizationProblem;
import wsi.ra.math.RNG;
import eva2.tools.EVAERROR;
import eva2.tools.Mathematics;
import eva2.tools.SelectedTag;
import wsi.ra.chart2d.DPoint;
import wsi.ra.chart2d.DPointSet;
import wsi.ra.math.Jama.Matrix;
/**
* This implements particle swarm optimization by Kennedy and Eberhardt.
@ -174,13 +172,13 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
protected void initIndividualDefaults(AbstractEAIndividual indy) {
double[] writeData;
// init velocity
writeData = new double[((InterfaceESIndividual)indy).getDGenotype().length];
writeData = new double[((InterfaceDataTypeDouble)indy).getDoubleData().length];
for (int j = 0; j < writeData.length; j++) {
writeData[j] = RNG.gaussianDouble(1.0);
//sum += (writeData[j])*(writeData[j]);
}
//sum = Math.sqrt(sum);
double relSpeed = getRelativeSpeed(writeData, ((InterfaceESIndividual)indy).getDoubleRange());
double relSpeed = getRelativeSpeed(writeData, ((InterfaceDataTypeDouble)indy).getDoubleRange());
for (int j = 0; j < writeData.length; j++) {
writeData[j] = (writeData[j]/relSpeed)*this.m_InitialVelocity;
}
@ -200,7 +198,7 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
System.arraycopy(tmpD, 0, writeData, 0, tmpD.length);
indy.SetData(partBestFitKey, writeData);
// init best position
tmpD = ((InterfaceESIndividual)indy).getDGenotype();
tmpD = ((InterfaceDataTypeDouble)indy).getDoubleData();
writeData = new double[tmpD.length];
System.arraycopy(tmpD, 0, writeData, 0, tmpD.length);
indy.SetData(partBestPosKey, writeData);
@ -212,11 +210,11 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
* @param population
*/
protected void traceEMA(Population population) {
if (population.get(0) instanceof InterfaceESIndividual) {
if (population.get(0) instanceof InterfaceDataTypeDouble) {
double[] curAvVelAndSpeed = getPopulationVelSpeed(population, 3);
double[][] range = ((InterfaceESIndividual)population.get(0)).getDoubleRange();
double[][] range = ((InterfaceDataTypeDouble)population.get(0)).getDoubleRange();
if (tracedVelocity == null) {
tracedVelocity = new double[((InterfaceESIndividual)population.get(0)).getDGenotype().length];
tracedVelocity = new double[((InterfaceDataTypeDouble)population.get(0)).getDoubleData().length];
for (int i=0; i<tracedVelocity.length; i++) tracedVelocity[i] = curAvVelAndSpeed[i];
} else {
if (population.getGeneration() < emaPeriods) {// if less than emaPeriods have passed, use larger alpha
@ -256,10 +254,10 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
*/
private double[] getPopulationVelSpeed(Population pop, int calcModeSwitch) {
AbstractEAIndividual indy = (AbstractEAIndividual)pop.get(0);
if (!(indy instanceof InterfaceESIndividual)) System.out.println("error, PSO needs ES individuals!");
if (!(indy instanceof InterfaceDataTypeDouble)) System.out.println("error, PSO needs individuals with double data!");
double[] ret;
double[][] range = ((InterfaceESIndividual)indy).getDoubleRange();
double[][] range = ((InterfaceDataTypeDouble)indy).getDoubleRange();
int retSize = 0;
///// warning, this method uses dark magic
@ -368,7 +366,7 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
// double sum;
for (int i = 0; i < this.m_Population.size(); i++) {
indy = (AbstractEAIndividual) this.m_Population.get(i);
if (indy instanceof InterfaceESIndividual) {
if (indy instanceof InterfaceDataTypeDouble) {
initIndividualDefaults(indy);
}
indy.SetData(indexKey, i);
@ -378,12 +376,13 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
for (int i = 0; i < this.m_Population.size(); i++) {
indy = (AbstractEAIndividual) this.m_Population.get(i);
if (indy instanceof InterfaceESIndividual) {
if (indy instanceof InterfaceDataTypeDouble) {
initIndividualMemory(indy);
}
}
this.m_BestIndividual = (AbstractEAIndividual)this.m_Population.getBestEAIndividual().clone();
if (reset) this.firePropertyChangedEvent("NextGenerationPerformed");
treeLevels = 0;
@ -438,7 +437,7 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
neighbourPos = (double[])neighbourIndy.getData(partBestPosKey);
} else {
neighbourFit = neighbourIndy.getFitness();
neighbourPos = ((InterfaceESIndividual)neighbourIndy).getDGenotype();
neighbourPos = ((InterfaceDataTypeDouble)neighbourIndy).getDoubleData();
}
if (neighbourFit == null || attractorFit == null) {
@ -453,15 +452,14 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
}
protected void resetIndividual(AbstractEAIndividual indy) {
if (indy instanceof InterfaceESIndividual) {
if (indy instanceof InterfaceDataTypeDouble) {
indy.setParents(null);
InterfaceESIndividual endy = (InterfaceESIndividual) indy;
endy.defaultInit();
indy.defaultInit();
indy.SetData(partTypeKey, defaultType); // turn into default type
initIndividualDefaults(indy);
initIndividualMemory(indy);
plotIndy(endy.getDGenotype(), null, (Integer)indy.getData(indexKey));
} else System.err.println("error, InterfaceESIndividual required (PSO)");
plotIndy(((InterfaceDataTypeDouble)indy).getDoubleData(), null, (Integer)indy.getData(indexKey));
} else System.err.println("error, double valued individuals required for PSO");
}
/** This method will update a given individual
@ -471,7 +469,7 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
* @param best The best individual found so far.
*/
protected void updateIndividual(int index, AbstractEAIndividual indy, Population pop) {
if (indy instanceof InterfaceESIndividual) {
if (indy instanceof InterfaceDataTypeDouble) {
int type=(Integer)indy.getData(partTypeKey);
switch (type) {
case resetType:
@ -488,13 +486,13 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
}
protected void defaultIndividualUpdate(int index, AbstractEAIndividual indy, Population pop) {
InterfaceESIndividual endy = (InterfaceESIndividual) indy;
InterfaceDataTypeDouble endy = (InterfaceDataTypeDouble) indy;
indy.SetData(partTypeKey, defaultType);
// default update
double[] personalBestPos = (double[]) indy.getData(partBestPosKey);
double[] velocity = (double[]) indy.getData(partVelKey);
double[] curPosition = endy.getDGenotype();
double[] curPosition = endy.getDoubleData();
double[][] range = endy.getDoubleRange();
// search for the local best position
@ -583,7 +581,7 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
*/
protected void updateIndProps(AbstractEAIndividual indy) {
indy.SetData(partBestFitKey, indy.getFitness());
indy.SetData(partBestPosKey, ((InterfaceESIndividual)indy).getDGenotype());
indy.SetData(partBestPosKey, ((InterfaceDataTypeDouble)indy).getDoubleData());
}
/**
@ -1005,7 +1003,7 @@ public class ParticleSwarmOptimization implements InterfaceOptimizer, java.io.Se
// finally set the new position and the current velocity
if (indy instanceof InterfaceDataTypeDouble) ((InterfaceDataTypeDouble)indy).SetDoubleGenotype(newPosition);
else {
((InterfaceESIndividual) indy).SetDGenotype(newPosition); // WARNING, this does a checkBounds in any case!
((InterfaceDataTypeDouble) indy).SetDoubleGenotype(newPosition); // WARNING, this does a checkBounds in any case!
if (!m_CheckConstraints) System.err.println("warning, checkbounds will be forced by InterfaceESIndividual!");
}

5
src/eva2/server/go/strategies/Tribes.java
View File

@ -7,7 +7,6 @@ import eva2.gui.GenericObjectEditor;
import eva2.server.go.InterfacePopulationChangedEventListener;
import eva2.server.go.individuals.AbstractEAIndividual;
import eva2.server.go.individuals.InterfaceDataTypeDouble;
import eva2.server.go.individuals.InterfaceESIndividual;
import eva2.server.go.populations.InterfaceSolutionSet;
import eva2.server.go.populations.Population;
import eva2.server.go.populations.SolutionSet;
@ -637,8 +636,8 @@ public class Tribes implements InterfaceOptimizer, java.io.Serializable {
public void setPopulation(Population pop) {
if (pop == null) return;
population = pop;
if (population.get(0) instanceof InterfaceESIndividual) {
range = ((InterfaceESIndividual)population.get(0)).getDoubleRange();
if (population.get(0) instanceof InterfaceDataTypeDouble) {
range = ((InterfaceDataTypeDouble)population.get(0)).getDoubleRange();
setDimension(range.length);
} else {
System.err.println("warning, TRIBES requires InterfaceESIndidivual instead of " + population.get(0).getClass() + ". Couldnt correctly init the problem range.");

43
src/eva2/server/go/strategies/tribes/TribesExplorer.java
View File

@ -8,7 +8,7 @@ import eva2.server.go.problems.InterfaceOptimizationProblem;
import eva2.server.go.strategies.Tribes;
import wsi.ra.math.RNG;
public class TribesExplorer extends AbstractEAIndividual implements InterfaceESIndividual, InterfaceDataTypeDouble {
public class TribesExplorer extends AbstractEAIndividual implements InterfaceDataTypeDouble {
/**
*
*/
@ -956,6 +956,17 @@ v[d] = cmin * v[d];
}
}
public void defaultInit() {
// shouldnt be called as we are beyond the EvA framework in this class
for (int i = 0; i < this.position.x.length; i++) {
this.position.x[0] = 0.;
}
}
public void defaultMutate() {
// shouldnt be called as we are beyond the EvA framework in this class
}
@Override
public void initByValue(Object obj, InterfaceOptimizationProblem opt) {
if (obj instanceof double[]) {
@ -1020,19 +1031,19 @@ v[d] = cmin * v[d];
return position.x.length;
}
public void SetDGenotype(double[] b) {
position.setDoubleArray(b);
}
public void defaultInit() {
System.err.println("defaultInit not available for TribesExplorer!");
}
public void defaultMutate() {
System.err.println("defaultMutate not available for TribesExplorer!");
}
public double[] getDGenotype() {
return position.getDoubleArray().clone();
}
// public void SetDGenotype(double[] b) {
// position.setDoubleArray(b);
// }
//
// public void defaultInit() {
// System.err.println("defaultInit not available for TribesExplorer!");
// }
//
// public void defaultMutate() {
// System.err.println("defaultMutate not available for TribesExplorer!");
// }
//
// public double[] getDGenotype() {
// return position.getDoubleArray().clone();
// }
}