001 /*
002 * Java Genetic Algorithm Library (jenetics-2.0.2).
003 * Copyright (c) 2007-2014 Franz Wilhelmstötter
004 *
005 * Licensed under the Apache License, Version 2.0 (the "License");
006 * you may not use this file except in compliance with the License.
007 * You may obtain a copy of the License at
008 *
009 * http://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 *
017 * Author:
018 * Franz Wilhelmstötter (franz.wilhelmstoetter@gmx.at)
019 */
020 package org.jenetics;
021
022 import static java.lang.String.format;
023 import static java.util.Objects.requireNonNull;
024
025 import java.util.Random;
026
027 import org.jenetics.internal.util.HashBuilder;
028
029 import org.jenetics.util.Factory;
030 import org.jenetics.util.RandomRegistry;
031
032 /**
033 * In tournament selection the best individual from a random sample of <i>s</i>
034 * individuals is chosen from the population <i>P<sub>g</sub></i>. The samples
035 * are drawn with replacement. An individual will win a tournament only if its
036 * fitness is greater than the fitness of the other <i>s-1</i> competitors.
037 * Note that the worst individual never survives, and the best individual wins
038 * in all the tournaments it participates. The selection pressure can be varied
039 * by changing the tournament size <i>s</i> . For large values of <i>s</i>, weak
040 * individuals have less chance being selected.
041 *
042 * @see <a href="http://en.wikipedia.org/wiki/Tournament_selection">Tournament selection</a>
043 *
044 * @author <a href="mailto:franz.wilhelmstoetter@gmx.at">Franz Wilhelmstötter</a>
045 * @since 1.0
046 * @version 2.0 — <em>$Date: 2014-08-12 $</em>
047 */
048 public class TournamentSelector<
049 G extends Gene<?, G>,
050 C extends Comparable<? super C>
051 >
052 implements Selector<G, C>
053 {
054
055 private final int _sampleSize;
056
057 /**
058 * Create a tournament selector with the give sample size. The sample size
059 * must be greater than one.
060 *
061 * @param sampleSize the number of individuals involved in one tournament
062 * @throws IllegalArgumentException if the sample size is smaller than two.
063 */
064 public TournamentSelector(final int sampleSize) {
065 if (sampleSize < 2) {
066 throw new IllegalArgumentException(
067 "Sample size must be greater than one, but was " + sampleSize
068 );
069 }
070 _sampleSize = sampleSize;
071 }
072
073 /**
074 * Create a tournament selector with sample size two.
075 */
076 public TournamentSelector() {
077 this(2);
078 }
079
080 @Override
081 public Population<G, C> select(
082 final Population<G, C> population,
083 final int count,
084 final Optimize opt
085 ) {
086 requireNonNull(population, "Population");
087 requireNonNull(opt, "Optimization");
088 if (count < 0) {
089 throw new IllegalArgumentException(format(
090 "Selection count must be greater or equal then zero, but was %s",
091 count
092 ));
093 }
094
095 final Population<G, C> pop = new Population<>(count);
096 final Factory<Phenotype<G, C>> factory = factory(
097 population, opt, _sampleSize, RandomRegistry.getRandom()
098 );
099
100 return pop.fill(factory, count);
101 }
102
103 private static <
104 G extends Gene<?, G>,
105 C extends Comparable<? super C>
106 >
107 Factory<Phenotype<G, C>> factory(
108 final Population<G, C> population,
109 final Optimize opt,
110 final int sampleSize,
111 final Random random
112 ) {
113 return new Factory<Phenotype<G, C>>() {
114 @Override
115 public Phenotype<G, C> newInstance() {
116 return select(population, opt, sampleSize, random);
117 }
118 };
119 }
120
121 private static <
122 G extends Gene<?, G>,
123 C extends Comparable<? super C>
124 >
125 Phenotype<G, C> select(
126 final Population<G, C> population,
127 final Optimize opt,
128 final int sampleSize,
129 final Random random
130 ) {
131 final int N = population.size();
132 Phenotype<G, C> winner = population.get(random.nextInt(N));
133
134 for (int j = 0; j < sampleSize; ++j) {
135 final Phenotype<G, C> selection = population.get(random.nextInt(N));
136 if (opt.compare(selection, winner) > 0) {
137 winner = selection;
138 }
139 }
140 assert (winner != null);
141
142 return winner;
143 }
144
145 @Override
146 public int hashCode() {
147 return HashBuilder.of(getClass()).and(_sampleSize).value();
148 }
149
150 @Override
151 public boolean equals(final Object obj) {
152 if (obj == this) {
153 return true;
154 }
155 if (obj == null || obj.getClass() != getClass()) {
156 return false;
157 }
158
159 final TournamentSelector<?, ?> selector = (TournamentSelector<?, ?>)obj;
160 return _sampleSize == selector._sampleSize;
161 }
162
163 @Override
164 public String toString() {
165 return format("%s[s=%d]", getClass().getSimpleName(), _sampleSize);
166 }
167
168 }
|