Distributed Genetic Programming Framework
print print

File org.dgpf.search.algorithms.ga.selection.OrderedSelection.java

Here you can find all the information about the file org.dgpf.search.algorithms.ga.selection.OrderedSelection.java. You may explore it here or download it onto your local disk.
 * Copyright (c) 2006 Thomas Weise
 * E-Mail           : tweise@gmx.de
 * Creation Date    : 2006-04-10 15:39:13
 * Original Filename: org.dgpf.search.algorithms.ga.selection.OrderedSelection.java
 * Version          : 2.0.2
 * Last modification: 2006-05-31
 *                by: Thomas Weise
 *                    Version 2.1, February 1999
 *                    You should have received a copy of this license along
 *                    with this library; if not, write to the Free Software
 *                    Foundation, Inc., 59 Temple Place, Suite 330, Boston,
 *                    MA 02111-1307, USA or download the license under
 *                    http://www.gnu.org/copyleft/lesser.html.
 * Warranty         : This software is provided "as is" 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.

package org.dgpf.search.algorithms.ga.selection;

import java.io.Serializable;

import org.dgpf.search.algorithms.ga.SelectionAlgorithm;
import org.dgpf.search.algorithms.ga.SortingAlgorithm;
import org.dgpf.search.api.Individual;
import org.sfc.math.stochastic.Randomizer;
import org.sfc.text.Text;

 * <p>
 * The ordered selection algorithm works sets the probability of an
 * individual being chosen for reproduction somehow related to its
 * rank in the population. Basically, all individuals are ordered by
 * their fitness. A uniform distributed random number is then raised to the
 * power of <code>o</code> and multplied with the list length. The
 * individual at the resulting index is then chosen for reproduction.
 * </p>
 * <p>
 * The selection algorithm parameter <code>o</code> represts the some
 * amount of overselection. The bigger it gets, the high the probability
 * of good individuals for being chosen for reproduction. An <code>o</code>
 * value of 1 means random selection, a value <code>0<=o<1</code> means
 * selecting only weak individuals. Values <code>o>1</code> favour
 * strong individuals. </p>
 * @see RankedSelection
 * @author Thomas Weise

public final class OrderedSelection extends SelectionAlgorithm
 * The serial version uid.

  private   static  final long    serialVersionUID        = 1;

 * A default instance of the tournament selection algorithms.

  public  static  final SelectionAlgorithm  DEFAULT_INSTANCE =
                                        new OrderedSelection(3);
 * The power to raise the random numbers to to calculate the index.

  private double  m_power ;

 * Create a new instance of the ordered selection.
 * @param p_power   The power to raise the random numbers to to calculate
 *                  the index.

  public  OrderedSelection  (final double p_power)
    this(p_power, null);
 * Create a new instance of the ordered selection.
 * @param p_power   The power to raise the random numbers to to calculate
 *                  the index.
 * @param p_sorting The underlying sorting algorithm.

  public  OrderedSelection  (final double           p_power,
                             final SortingAlgorithm p_sorting)
    this.m_power = p_power;
 * Select one individual out of the population for reproduction (crossover
 * or mutation). The individuals in the array are ordered by the search's
 * current comparator, with the fittest individual at index <code>0</code>
 * and the worst at index <code>p_count-1</code>.
 * @param p_random        The randomizer to be used for the selection.
 * @param p_individuals   The population of individuals after being
 *                        preprocessed.
 * @param p_count         The count of valid items inside the population
 *                        array.
 * @param <Genotype>      The genotype of the individuals to select.
 * @return  An individual to reproduce.

  public  final  <Genotype extends Serializable> Individual<Genotype>
                         select(final Randomizer             p_random,
                                final Individual<Genotype>[] p_individuals,
                                final int                    p_count)
    return p_individuals[(int)(Math.pow(p_random.nextDouble(), this.m_power)
                         * p_count)];

 * Returns the name of the genetic selection algorithm.
 * @return The name of the genetic selection algorithm.

  public  final String  toString  ()
    return "Ordered Selection (power=" +
           Text.double_to_string(this.m_power) + ')';

 * Check whether this object equals another one. This method is internally
 * invoked by <code>equals()</code>. 
 * @param p_object  The object to compare with.
 * @return <code>true</code> if and only if this object equals to the other
 *         one.

  protected final boolean is_equal  (final Object p_object)
    return ((p_object instanceof OrderedSelection) &&
                this.m_power) == 0));

File Information:

file name:OrderedSelection.java
qualified name:org.dgpf.search.algorithms.ga.selection.OrderedSelection.java
file type:Java Source File
download location:download http://dgpf.sourceforge.net/source/org/dgpf/search/algorithms/ga/selection/OrderedSelection.java
size:5.318 KB (5446 B)
uploaded: 2018-01-07 12:03:35 GMT+0000
last update: 2006-05-31 03:15:08 GMT+0000
last access: 2018-05-25 15:06:32 GMT+0000

statistics online since 2006-01-02.   RSS Feed
Contact us by sending an email to tweise@gmx.de to receive further information, to report errors, or to join our project.
All content on this site (http://dgpf.sourceforge.net/) is LGPL-licensed.
http://dgpf.sourceforge.net/scripts/source/source.php last modified at 2018-01-07 12:03:34 GMT+0000 served at 2018-05-25 15:06:32 GMT+0000.
Valid CSS Valid XHTML 1.1
Valid RSS SourceForge.net Logo