nojhan/paradiseo
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Ajout de la random neutral walk sampling, on tient le bon bout ;)

Browse source 
git-svn-id: svn://scm.gforge.inria.fr/svnroot/paradiseo@1797 331e1502-861f-0410-8da2-ba01fb791d7f
This commit is contained in:
verel 2010-05-07 09:05:31 +00:00
commit 312b213a01
8 changed files with 614 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
trunk/paradiseo-mo/src/sampling/moMHRndFitnessCloudSampling.h
View file

@ -52,7 +52,7 @@
* Sample the fitness of solutions from Metropolis-Hasting sampling
* and the fitness of one random neighbor
*
* The values are collected during the random search
* The values are collected during the Metropolis-Hasting walk
*
*/
template <class Neighbor>

148
trunk/paradiseo-mo/src/sampling/moNeutralWalkSampling.h Normal file
View file

@ -0,0 +1,148 @@
/*
<moNeutralWalkSampling.h>
Copyright (C) DOLPHIN Project-Team, INRIA Lille - Nord Europe, 2006-2010
Sebastien Verel, Arnaud Liefooghe, Jeremie Humeau
This software is governed by the CeCILL license under French law and
abiding by the rules of distribution of free software. You can use,
modify and/ or redistribute the software under the terms of the CeCILL
license as circulated by CEA, CNRS and INRIA at the following URL
"http://www.cecill.info".
As a counterpart to the access to the source code and rights to copy,
modify and redistribute granted by the license, users are provided only
with a limited warranty and the software's author, the holder of the
economic rights, and the successive licensors have only limited liability.
In this respect, the user's attention is drawn to the risks associated
with loading, using, modifying and/or developing or reproducing the
software by the user in light of its specific status of free software,
that may mean that it is complicated to manipulate, and that also
therefore means that it is reserved for developers and experienced
professionals having in-depth computer knowledge. Users are therefore
encouraged to load and test the software's suitability as regards their
requirements in conditions enabling the security of their systems and/or
data to be ensured and, more generally, to use and operate it in the
same conditions as regards security.
The fact that you are presently reading this means that you have had
knowledge of the CeCILL license and that you accept its terms.
ParadisEO WebSite : http://paradiseo.gforge.inria.fr
Contact: paradiseo-help@lists.gforge.inria.fr
*/
#ifndef moNeutralWalkSampling_h
#define moNeutralWalkSampling_h
#include <eoInit.h>
#include <eval/moEval.h>
#include <eoEvalFunc.h>
#include <algo/moRandomNeutralWalk.h>
#include <sampling/moSampling.h>
#include <perturb/moSolInit.h>
#include <continuator/moSolutionStat.h>
#include <utils/eoDistance.h>
#include <continuator/moDistanceStat.h>
#include <continuator/moNeighborhoodStat.h>
#include <continuator/moMaxNeighborStat.h>
#include <continuator/moMinNeighborStat.h>
#include <continuator/moAverageFitnessNeighborStat.h>
#include <continuator/moStdFitnessNeighborStat.h>
#include <continuator/moSizeNeighborStat.h>
#include <continuator/moNbInfNeighborStat.h>
#include <continuator/moNbSupNeighborStat.h>
#include <continuator/moNeutralDegreeNeighborStat.h>
/**
* To explore the evolvability of solutions in a neutral networks:
* Perform a random neutral walk based on the neighborhood,
* The measures of evolvability of solutions are collected during the random neutral walk
* The distribution and autocorrelation can be computed from the serie of values
*
* Informations collected:
* - the current solution of the walk
* - the distance from the starting solution
* - the minimal fitness in the neighborhood
* - the average fitness
* - the standard deviation of the fitness
* - the maximal fitness
* - the size of the neighborhood
* - the number of neighbors with lower fitness
* - the number of neighbors with equal fitness (neutral degree)
* - the number of neighbors with higher fitness
*/
template <class Neighbor>
class moNeutralWalkSampling : public moSampling<Neighbor>
{
public:
typedef typename Neighbor::EOT EOT ;
using moSampling<Neighbor>::localSearch;
/**
* Default Constructor
* @param _initSol the first the solution of the walk
* @param _neighborhood neighborhood giving neighbor in random order
* @param _fullEval Fitness function, full evaluation function
* @param _eval neighbor evaluation, incremental evaluation function
* @param _distance the distance to measure the distance from the initial solution
* @param _nbStep Number of steps of the random walk
*/
moNeutralWalkSampling(EOT & _initSol,
moNeighborhood<Neighbor> & _neighborhood,
eoEvalFunc<EOT>& _fullEval,
moEval<Neighbor>& _eval,
eoDistance<EOT> & _distance,
unsigned int _nbStep) :
moSampling<Neighbor>(init, * new moRandomNeutralWalk<Neighbor>(_neighborhood, _fullEval, _eval, _nbStep), solutionStat),
init(_initSol),
distStat(_distance, _initSol),
neighborhoodStat(_neighborhood, _eval),
minStat(neighborhoodStat),
averageStat(neighborhoodStat),
stdStat(neighborhoodStat),
maxStat(neighborhoodStat),
nbSupStat(neighborhoodStat),
nbInfStat(neighborhoodStat),
sizeStat(neighborhoodStat),
ndStat(neighborhoodStat)
{
add(neighborhoodStat, false);
add(distStat);
add(minStat);
add(averageStat);
add(stdStat);
add(maxStat);
add(sizeStat);
add(nbInfStat);
add(ndStat);
add(nbSupStat);
}
/**
* default destructor
*/
~moNeutralWalkSampling() {
// delete the pointer on the local search which has been constructed in the constructor
delete localSearch;
}
protected:
moSolInit<EOT> init;
moSolutionStat<EOT> solutionStat;
moDistanceStat<EOT> distStat;
moNeighborhoodStat< Neighbor > neighborhoodStat;
moMinNeighborStat< Neighbor > minStat;
moAverageFitnessNeighborStat< Neighbor > averageStat;
moStdFitnessNeighborStat< Neighbor > stdStat;
moMaxNeighborStat< Neighbor > maxStat;
moNbSupNeighborStat< Neighbor > nbSupStat;
moNbInfNeighborStat< Neighbor > nbInfStat;
moSizeNeighborStat< Neighbor > sizeStat;
moNeutralDegreeNeighborStat< Neighbor > ndStat;
};
#endif