001 /*
002 * Java Genetic Algorithm Library (jenetics-6.2.0).
003 * Copyright (c) 2007-2021 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@gmail.com)
019 */
020 package io.jenetics;
021
022 import static java.lang.String.format;
023 import static java.util.Objects.requireNonNull;
024
025 import java.util.Comparator;
026 import java.util.Random;
027 import java.util.stream.Stream;
028
029 import io.jenetics.util.ISeq;
030 import io.jenetics.util.MSeq;
031 import io.jenetics.util.RandomRegistry;
032 import io.jenetics.util.Seq;
033
034 /**
035 * In tournament selection the best individual from a random sample of <i>s</i>
036 * individuals is chosen from the population <i>P<sub>g</sub></i>. The samples
037 * are drawn with replacement. An individual will win a tournament only if its
038 * fitness is greater than the fitness of the other <i>s-1</i> competitors.
039 * Note that the worst individual never survives, and the best individual wins
040 * in all the tournaments it participates. The selection pressure can be varied
041 * by changing the tournament size <i>s</i> . For large values of <i>s</i>, weak
042 * individuals have less chance being selected.
043 *
044 * @see <a href="http://en.wikipedia.org/wiki/Tournament_selection">Tournament selection</a>
045 *
046 * @author <a href="mailto:franz.wilhelmstoetter@gmail.com">Franz Wilhelmstötter</a>
047 * @since 1.0
048 * @version 6.0
049 */
050 public class TournamentSelector<
051 G extends Gene<?, G>,
052 C extends Comparable<? super C>
053 >
054 implements Selector<G, C>
055 {
056
057 private final Comparator<? super Phenotype<G, C>> _comparator;
058 private final int _sampleSize;
059
060 /**
061 * Create a tournament selector with the give {@code comparator} and
062 * sample size. The sample size must be greater than one.
063 *
064 * @since 6.0
065 *
066 * @param comparator the comparator use for comparing two individuals during
067 * a tournament
068 * @param sampleSize the number of individuals involved in one tournament
069 * @throws IllegalArgumentException if the sample size is smaller than two
070 * @throws NullPointerException if the given {@code comparator} is
071 * {@code null}
072 */
073 public TournamentSelector(
074 final Comparator<? super Phenotype<G, C>> comparator,
075 final int sampleSize
076 ) {
077 _comparator = requireNonNull(comparator);
078 if (sampleSize < 2) {
079 throw new IllegalArgumentException(
080 "Sample size must be greater than one, but was " + sampleSize
081 );
082 }
083 _sampleSize = sampleSize;
084 }
085
086 /**
087 * Create a tournament selector with the give sample size. The sample size
088 * must be greater than one.
089 *
090 * @param sampleSize the number of individuals involved in one tournament
091 * @throws IllegalArgumentException if the sample size is smaller than two.
092 */
093 public TournamentSelector(final int sampleSize) {
094 this(Phenotype::compareTo, sampleSize);
095 }
096
097 /**
098 * Create a tournament selector with sample size two.
099 */
100 public TournamentSelector() {
101 this(Phenotype::compareTo,2);
102 }
103
104 /**
105 * Return the sample size of the tournament selector.
106 *
107 * @since 5.0
108 *
109 * @return the sample size of the tournament selector
110 */
111 public int sampleSize() {
112 return _sampleSize;
113 }
114
115 @Override
116 public ISeq<Phenotype<G, C>> select(
117 final Seq<Phenotype<G, C>> population,
118 final int count,
119 final Optimize opt
120 ) {
121 requireNonNull(population, "Population");
122 requireNonNull(opt, "Optimization");
123 if (count < 0) {
124 throw new IllegalArgumentException(format(
125 "Selection count must be greater or equal then zero, but was %s",
126 count
127 ));
128 }
129
130 final Random random = RandomRegistry.random();
131 return population.isEmpty()
132 ? ISeq.empty()
133 : MSeq.<Phenotype<G, C>>ofLength(count)
134 .fill(() -> select(population, opt, random))
135 .toISeq();
136 }
137
138 private Phenotype<G, C> select(
139 final Seq<Phenotype<G, C>> population,
140 final Optimize opt,
141 final Random random
142 ) {
143 final int N = population.size();
144
145 assert _sampleSize >= 2;
146 assert N >= 1;
147
148 final Comparator<? super Phenotype<G, C>> cmp = opt == Optimize.MAXIMUM
149 ? _comparator
150 : _comparator.reversed();
151
152 return Stream.generate(() -> population.get(random.nextInt(N)))
153 .limit(_sampleSize)
154 .max(cmp)
155 .orElseThrow(AssertionError::new);
156 }
157
158 @Override
159 public String toString() {
160 return format("%s[s=%d]", getClass().getSimpleName(), _sampleSize);
161 }
162
163 }
|