001/*
002 * Java Genetic Algorithm Library (jenetics-8.1.0).
003 * Copyright (c) 2007-2024 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 */
020package io.jenetics;
021
022import static java.lang.Math.min;
023import static java.lang.String.format;
024import static java.util.Objects.requireNonNull;
025
026import io.jenetics.util.ISeq;
027import io.jenetics.util.MSeq;
028import io.jenetics.util.Seq;
029
030/**
031 * In truncation selection, individuals are sorted according to their fitness.
032 * Only the n  best individuals are selected. The truncation selection is a very
033 * basic selection algorithm. It has its strength in fast selecting individuals
034 * in large populations, but is seldom used in practice.
035 *
036 * @see <a href="http://en.wikipedia.org/wiki/Truncation_selection">
037 *          Wikipedia: Truncation selection
038 *      </a>
039 *
040 * @author <a href="mailto:franz.wilhelmstoetter@gmail.com">Franz Wilhelmstötter</a>
041 * @since 1.0
042 * @version 5.0
043 */
044public final class TruncationSelector<
045        G extends Gene<?, G>,
046        C extends Comparable<? super C>
047>
048        implements Selector<G, C>
049{
050
051        private final int _n;
052
053        /**
054         * Create a new {@code TruncationSelector} object, where the worst selected
055         * individual has rank {@code n}. This means, if you want to select
056         * {@code count} individuals, the worst selected individual has rank
057         * {@code n}. If {@code count > n}, the selected population will contain
058         * <em>duplicate</em> individuals.
059         *
060         * @since 3.8
061         *
062         * @param n the worst rank of the selected individuals
063         * @throws IllegalArgumentException if {@code n < 1}
064         */
065        public TruncationSelector(final int n) {
066                if (n < 1) {
067                        throw new IllegalArgumentException(format(
068                                "n must be greater or equal 1, but was %d.", n
069                        ));
070                }
071
072                _n = n;
073        }
074
075        /**
076         * Create a new TruncationSelector object.
077         */
078        public TruncationSelector() {
079                this(Integer.MAX_VALUE);
080        }
081
082        /**
083         * This method sorts the population in descending order while calculating
084         * the selection probabilities. If the selection size is greater the
085         * population size, the whole population is duplicated until the desired
086         * sample size is reached.
087         *
088         * @throws NullPointerException if the {@code population} or {@code opt} is
089         *         {@code null}.
090         */
091        @Override
092        public ISeq<Phenotype<G, C>> select(
093                final Seq<Phenotype<G, C>> population,
094                final int count,
095                final Optimize opt
096        ) {
097                requireNonNull(population, "Population");
098                requireNonNull(opt, "Optimization");
099                if (count < 0) {
100                        throw new IllegalArgumentException(format(
101                                "Selection count must be greater or equal then zero, but was %s",
102                                count
103                        ));
104                }
105
106                final MSeq<Phenotype<G, C>> selection = MSeq
107                        .ofLength(population.isEmpty() ? 0 : count);
108
109                if (count > 0 && !population.isEmpty()) {
110                        final MSeq<Phenotype<G, C>> copy = population.asISeq().copy();
111                        copy.sort((a, b) ->
112                                opt.<C>descending().compare(a.fitness(), b.fitness()));
113
114                        int size = count;
115                        do {
116                                final int length = min(min(copy.size(), size), _n);
117                                for (int i = 0; i < length; ++i) {
118                                        selection.set((count - size) + i, copy.get(i));
119                                }
120
121                                size -= length;
122                        } while (size > 0);
123                }
124
125                return selection.toISeq();
126        }
127
128        @Override
129        public String toString() {
130                return getClass().getName();
131        }
132
133}