/*
* Copyright 2007-2013
* Licensed under GNU Lesser General Public License
*
* This file is part of EpochX
*
* EpochX is free software: you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* EpochX 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 Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with EpochX. If not, see <http://www.gnu.org/licenses/>.
*
* The latest version is available from: http:/www.epochx.org
*/
package org.epochx.cfg.benchmark;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import org.epochx.Breeder;
import org.epochx.Config.ConfigKey;
import org.epochx.BranchedBreeder;
import org.epochx.EvolutionaryStrategy;
import org.epochx.FitnessEvaluator;
import org.epochx.GenerationalStrategy;
import org.epochx.GenerationalTemplate;
import org.epochx.Initialiser;
import org.epochx.MaximumGenerations;
import org.epochx.Operator;
import org.epochx.Population;
import org.epochx.RandomSequence;
import org.epochx.TerminationCriteria;
import org.epochx.TerminationFitness;
import org.epochx.cfg.CFGIndividual;
import org.epochx.cfg.CFGSourceGenerator;
import org.epochx.cfg.fitness.CFGFitnessFunction;
import org.epochx.cfg.fitness.HitsCount;
import org.epochx.cfg.init.Grow;
import org.epochx.cfg.operator.SubtreeCrossover;
import org.epochx.cfg.operator.SubtreeMutation;
import org.epochx.fitness.DoubleFitness;
import org.epochx.grammar.Grammar;
import org.epochx.interpret.EpoxInterpreter;
import org.epochx.random.MersenneTwisterFast;
import org.epochx.selection.TournamentSelector;
import org.epochx.tools.BenchmarkSolutions;
/**
* This template sets up EpochX to run the cubic regression benchmark with
* the GE representation. Cubic regression involves evolving an equivalent
* function to the formula: x + x^2 + x^3
*
* The following configuration is used:
*
* <ul>
* <li>{@link Population#SIZE}: <code>100</code>
* <li>{@link GenerationalStrategy#TERMINATION_CRITERIA}: <code>MaximumGenerations</code>, <code>TerminationFitness(0.0)</code>
* <li>{@link MaximumGenerations#MAXIMUM_GENERATIONS}: <code>50</code>
* <li>{@link GEIndividual#MAXIMUM_DEPTH}: <code>17</code>
* <li>{@link BranchedBreeder#SELECTOR}: <code>TournamentSelector</code>
* <li>{@link TournamentSelector#TOURNAMENT_SIZE}: <code>7</code>
* <li>{@link Breeder#OPERATORS}: <code>OnePointCrossover</code>, <code>PointMutation</code>
* <li>{@link OnePointCrossover#PROBABILITY}: <code>0.0</code>
* <li>{@link PointMutation#PROBABILITY}: <code>1.0</code>
* <li>{@link Initialiser#METHOD}: <code>GrowInitialiser</code>
* <li>{@link RandomSequence#RANDOM_SEQUENCE}: <code>MersenneTwisterFast</code>
* <li>{@link Grammar#GRAMMER}: [Listed below]
* <li>{@link CodonFactory#CODON_FACTORY}: <code>IntegerCodonFactory</code>
* <li>{@link CFGFitnessFunction#INTERPRETER}: <code>EpoxInterpreter(GESourceGenerator)</code>
* <li>{@link MappingComponent#MAPPER}: <code>DepthFirstMapper</code>
* <li>{@link FitnessEvaluator#FUNCTION}: <code>HitsCount</code>
* <li>{@link HitsCount#POINT_ERROR}: <code>0.01</code>
* <li>{@link HitsCount#INPUT_IDENTIFIERS}: <code>new String[]{"X"}</code>
* <li>{@link HitsCount#INPUT_VALUE_SETS}: [20 random values between -1.0 and +1.0]
* <li>{@link HitsCount#EXPECTED_OUTPUTS}: [correct output for input value sets]
*
* <h3>Grammar</h3>
*
* {@code
* <prog> ::= <node>
* <node> ::= <function> | <terminal>
* <function> ::= ADD( <node> , <node> )
* | SUB( <node> , <node> )
* | MUL( <node> , <node> )
* | DIV( <node> , <node> )
* <terminal> ::= X
*
* @since 2.0
* }
*/
public class CubicRegression extends GenerationalTemplate {
/**
* Sets up the given template with the benchmark config settings
*
* @param template a map to be filled with the template config
*/
@Override
protected void fill(Map<ConfigKey<?>, Object> template) {
super.fill(template);
template.put(Population.SIZE, 100);
List<TerminationCriteria> criteria = new ArrayList<TerminationCriteria>();
criteria.add(new TerminationFitness(new DoubleFitness.Minimise(0.0)));
criteria.add(new MaximumGenerations());
template.put(EvolutionaryStrategy.TERMINATION_CRITERIA, criteria);
template.put(MaximumGenerations.MAXIMUM_GENERATIONS, 50);
template.put(CFGIndividual.MAXIMUM_DEPTH, 17);
template.put(Breeder.SELECTOR, new TournamentSelector());
template.put(TournamentSelector.TOURNAMENT_SIZE, 7);
List<Operator> operators = new ArrayList<Operator>();
operators.add(new SubtreeCrossover());
operators.add(new SubtreeMutation());
template.put(Breeder.OPERATORS, operators);
template.put(SubtreeCrossover.PROBABILITY, 0.0);
template.put(SubtreeMutation.PROBABILITY, 1.0);
template.put(Initialiser.METHOD, new Grow());
RandomSequence randomSequence = new MersenneTwisterFast();
template.put(RandomSequence.RANDOM_SEQUENCE, randomSequence);
// Setup grammar
String grammarStr = "<prog> ::= <node>\n"
+ "<node> ::= <function> | <terminal>\n"
+ "<function> ::= ADD( <node> , <node> ) "
+ "| SUB( <node> , <node> ) "
+ "| MUL( <node> , <node> ) "
+ "| DIV( <node> , <node> )\n"
+ "<terminal> ::= X\n";
Grammar grammar = new Grammar(grammarStr);
template.put(Grammar.GRAMMAR, grammar);
template.put(CFGFitnessFunction.INTERPRETER, new EpoxInterpreter<CFGIndividual>(new CFGSourceGenerator()));
// Generate inputs and expected outputs
int noPoints = 20;
Double[][] inputs = new Double[noPoints][1];
Double[] expectedOutputs = new Double[noPoints];
for (int i=0; i<noPoints; i++) {
// Inputs values between -1.0 and +1.0
inputs[i][0] = (randomSequence.nextDouble() * 2) - 1;
expectedOutputs[i] = BenchmarkSolutions.cubicRegression(inputs[i][0]);
}
// Setup fitness function
template.put(FitnessEvaluator.FUNCTION, new HitsCount());
template.put(HitsCount.POINT_ERROR, 0.01);
template.put(HitsCount.INPUT_IDENTIFIERS, new String[]{"X"});
template.put(HitsCount.INPUT_VALUE_SETS, inputs);
template.put(HitsCount.EXPECTED_OUTPUTS, expectedOutputs);
}
}