cstrs_core.cpp

/*****************************************************************************
 *   Copyright (C) 2004-2015 The PaGMO development team,                     *
 *   Advanced Concepts Team (ACT), European Space Agency (ESA)               *
 *                                                                           *
 *   https://github.com/esa/pagmo                                            *
 *                                                                           *
 *   act@esa.int                                                             *
 *                                                                           *
 *   This program is free software; you can redistribute it and/or modify    *
 *   it under the terms of the GNU General Public License as published by    *
 *   the Free Software Foundation; either version 2 of the License, or       *
 *   (at your option) any later version.                                     *
 *                                                                           *
 *   This program is distributed in the hope that it will be useful,         *
 *   but WITHOUT ANY WARRANTY; without even the implied warranty of          *
 *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the           *
 *   GNU General Public License for more details.                            *
 *                                                                           *
 *   You should have received a copy of the GNU General Public License       *
 *   along with this program; if not, write to the                           *
 *   Free Software Foundation, Inc.,                                         *
 *   59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.               *
 *****************************************************************************/

#include <boost/random/uniform_int.hpp>
#include <boost/random/uniform_real.hpp>
#include <boost/random/variate_generator.hpp>
#include <boost/random/normal_distribution.hpp>
#include <string>
#include <vector>

#include "../exceptions.h"
#include "../population.h"
#include "../problem/base.h"
#include "../problem/con2uncon.h"
#include "../types.h"
#include "base.h"
#include "cstrs_core.h"

namespace pagmo { namespace algorithm {


cstrs_core::cstrs_core(const base &original_algo, const base &repair_algo,
                       int gen,
                                           int repair_frequency,
                                           double repair_ratio,
                                           double ftol, double xtol):
    base(),m_original_algo(original_algo.clone()),
    m_repair_algo(repair_algo.clone()),
    m_gen(gen),m_repair_frequency(repair_frequency),
        m_repair_ratio(repair_ratio),m_ftol(ftol),m_xtol(xtol)
{
//      m_original_algo = original_algo.clone();
//    m_repair_algo = repair_algo.clone();

        if(gen < 0) {
                pagmo_throw(value_error,"number of generations must be nonnegative");
        }
        if(repair_frequency < 0) {
                pagmo_throw(value_error,"repair frequency must be positive");
        }
        if((repair_ratio < 0) || (repair_ratio > 1)) {
                pagmo_throw(value_error,"repair ratio must be in [0..1]");
        }
}

cstrs_core::cstrs_core(const cstrs_core &algo):
    base(algo),m_original_algo(algo.m_original_algo->clone()),
    m_repair_algo(algo.m_repair_algo->clone()),m_gen(algo.m_gen),
    m_repair_frequency(algo.m_repair_frequency),
    m_repair_ratio(algo.m_repair_ratio),
    m_ftol(algo.m_ftol),m_xtol(algo.m_xtol)
{}

base_ptr cstrs_core::clone() const
{
        return base_ptr(new cstrs_core(*this));
}


void cstrs_core::evolve(population &pop) const
{       
        // store useful variables
        const problem::base &prob = pop.problem();
        const population::size_type pop_size = pop.size();
        const problem::base::size_type prob_dimension = prob.get_dimension();

        // get the constraints dimension
        problem::base::c_size_type prob_c_dimension = prob.get_c_dimension();

        //We perform some checks to determine wether the problem/population are suitable for CORE
        if(prob_c_dimension < 1) {
                pagmo_throw(value_error,"The problem is not constrained and CORE is not suitable to solve it");
        }
        if(prob.get_f_dimension() != 1) {
                pagmo_throw(value_error,"The problem is multiobjective and CORE is not suitable to solve it");
        }

        // Get out if there is nothing to do.
        if(pop_size == 0) {
                return;
        }

        // generates the unconstrained problem
        problem::con2uncon prob_unconstrained(prob);

        // associates the population to this problem
        population pop_uncon(prob_unconstrained);

        // fill this unconstrained population
        pop_uncon.clear();
        for(population::size_type i=0; i<pop_size; i++) {
                pop_uncon.push_back(pop.get_individual(i).cur_x);
        }

        // vector containing the infeasibles positions
        std::vector<population::size_type> pop_infeasibles;

        // Main CORE loop
        for(int k=0; k<m_gen; k++) {

                if(k%m_repair_frequency == 0) {
                        pop_infeasibles.clear();

                        // get the infeasible individuals
                        for(population::size_type i=0; i<pop_size; i++) {
                                if(!prob.feasibility_c(pop.get_individual(i).cur_c)) {
                                        pop_infeasibles.push_back(i);
                                }
                        }

                        // random shuffle of infeasibles?
                        population::size_type number_of_repair = (population::size_type)(m_repair_ratio * pop_infeasibles.size());

                        // repair the infeasible individuals
                        for(population::size_type i=0; i<number_of_repair; i++) {
                                const population::size_type &current_individual_idx = pop_infeasibles.at(i);

                pop.repair(current_individual_idx, m_repair_algo);
                        }

                        // the population is repaired, it can be now used in the new unconstrained population
                        // only the repaired individuals are put back in the population
                        for(population::size_type i=0; i<number_of_repair; i++) {
                                population::size_type current_individual_idx = pop_infeasibles.at(i);
                                pop_uncon.set_x(current_individual_idx, pop.get_individual(current_individual_idx).cur_x);
                        }
                }

                m_original_algo->evolve(pop_uncon);

                // push back the population in the main problem
                pop.clear();
                for(population::size_type i=0; i<pop_size; i++) {
                        pop.push_back(pop_uncon.get_individual(i).cur_x);
                }

                // Check the exit conditions (every 40 generations, just as DE)
                if(k % 40 == 0) {
                        decision_vector tmp(prob_dimension);

                        double dx = 0;
                        for(decision_vector::size_type i=0; i<prob_dimension; i++) {
                                tmp[i] = pop.get_individual(pop.get_worst_idx()).best_x[i] - pop.get_individual(pop.get_best_idx()).best_x[i];
                                dx += std::fabs(tmp[i]);
                        }

                        if(dx < m_xtol ) {
                                if (m_screen_output) {
                                        std::cout << "Exit condition -- xtol < " << m_xtol << std::endl;
                                }
                                break;
                        }

                        double mah = std::fabs(pop.get_individual(pop.get_worst_idx()).best_f[0] - pop.get_individual(pop.get_best_idx()).best_f[0]);

                        if(mah < m_ftol) {
                                if(m_screen_output) {
                                        std::cout << "Exit condition -- ftol < " << m_ftol << std::endl;
                                }
                                break;
                        }

                        // outputs current values
                        if(m_screen_output) {
                                std::cout << "Generation " << k << " ***" << std::endl;
                                std::cout << "    Best global fitness: " << pop.champion().f << std::endl;
                                std::cout << "    xtol: " << dx << ", ftol: " << mah << std::endl;
                                std::cout << "    xtol: " << dx << ", ftol: " << mah << std::endl;
                        }
                }
        }
}

std::string cstrs_core::get_name() const
{
        return m_original_algo->get_name() + "[CORE]";
}


base_ptr cstrs_core::get_algorithm() const
{
        return m_original_algo->clone();
}


void cstrs_core::set_algorithm(const base &algo)
{
        m_original_algo = algo.clone();
}


base_ptr cstrs_core::get_repair_algorithm() const
{
    return m_repair_algo->clone();
}


void cstrs_core::set_repair_algorithm(const base &algo)
{
    m_repair_algo = algo.clone();
}


std::string cstrs_core::human_readable_extra() const
{
        std::ostringstream s;
        s << "algorithms: " << m_original_algo->get_name() << " ";
        s << "\n\tConstraints handled with CORE algorithm";
        return s.str();
}

}} //namespaces

BOOST_CLASS_EXPORT_IMPLEMENT(pagmo::algorithm::cstrs_core)