CMAPlot.cpp

Go to the documentation of this file.

/*!
 * 
 *
 * \brief       Example for examining characteristics of the CMA with the help of
 * the Probe framework.
 * 
 * 
 *
 * \author      tvoss
 * \date        -
 *
 *
 * \par Copyright 1995-2017 Shark Development Team
 * 
 * <BR><HR>
 * This file is part of Shark.
 * <https://shark-ml.github.io/Shark/>
 * 
 * Shark is free software: you can redistribute it and/or modify
 * it under the terms of the GNU Lesser General Public License as published 
 * by the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 * 
 * Shark 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 Lesser General Public License for more details.
 * 
 * You should have received a copy of the GNU Lesser General Public License
 * along with Shark.  If not, see <http://www.gnu.org/licenses/>.
 *
 */
 
#include <shark/Algorithms/DirectSearch/CMA.h>
#include <shark/ObjectiveFunctions/Benchmarks/Benchmarks.h>
 
using namespace shark;
using namespace std;
 
#include <boost/property_tree/json_parser.hpp>
 
int main( int argc, char ** argv ) {
 
    // Results go here.
    ofstream results( "results.txt" );
    // Plotting commands (gnuplot) go here.
    ofstream plot( "plot.txt" );
    plot << "set key outside bottom center" << endl;
    plot << "set size square" << endl;
    plot << "set zeroaxis" << endl;
    plot << "set border 0" << endl;
    plot << "set xrange [-4:4]" << endl;
    plot << "set yrange [-4:4]" << endl;
    // Adjust the floating-point format to scientific and increase output precision.
    results.setf( ios_base::scientific );
    results.precision( 10 );
    plot.setf( ios_base::scientific );
    plot.precision( 10 );
 
    // Instantiate both the problem and the optimizer.
    benchmarks::Himmelblau hb;
    CMA cma;
    hb.init();
    cma.init( hb );
 
    // Iterate the optimizer until a solution of sufficient quality is found.
    do{
        // Print error ellipses for covariance matrices.
        plot << "set object " 
             << hb.evaluationCounter() + 1
             << " ellipse center " 
             << cma.mean()( 0 ) << ","
             << cma.mean()( 1 ) << " size "
             << cma.eigenValues()(0) * cma.sigma() * 2. << "," // times 2 because gnuplot takes diameters as arguments 
             << cma.eigenValues()(1) * cma.sigma() * 2. << " angle " 
             << ::atan( cma.eigenVectors()( 1, 0 ) / cma.eigenVectors()( 0, 0 ) ) / M_PI * 180 << " front fillstyle empty border 2" << endl;
        
        // Report information on the optimizer state and the current solution to the console.
        results << hb.evaluationCounter() << " "    // Column 1
            << cma.condition() << " "       // Column 2
            << cma.sigma() << " "           // Column 3
            << cma.solution().value << " ";     // Column 4
        copy(
             cma.solution().point.begin(),              
             cma.solution().point.end(),                // Column 5 & 6
             ostream_iterator< double >( results, " " ) 
             );
        copy( 
             cma.mean().begin(),                        // Column 7 & 8
             cma.mean().end(), 
             ostream_iterator< double >( results, " " ) 
              );
        results << endl;
 
        // Do one CMA iteration/generation.
        cma.step( hb );
 
    } while( cma.solution().value> 1E-20 );
 
    // Write final result.
    // Print error ellipses for covariance matrices.
    plot << "set object " 
         << hb.evaluationCounter() + 1
         << " ellipse center " 
         << cma.mean()( 0 ) << ","
         << cma.mean()( 1 ) << " size "
         << cma.eigenValues()(0) * cma.sigma() * 2. << "," // times 2 because gunplot takes diameters as arguments 
         << cma.eigenValues()(1) * cma.sigma() * 2. << " angle " 
         << ::atan( cma.eigenVectors()( 1, 0 ) / cma.eigenVectors()( 0, 0 ) ) / M_PI * 180  << " front fillstyle empty border 2" << endl;
    
    // Report information on the optimizer state and the current solution to the console.
    results << hb.evaluationCounter() << " "    // Column 1
        << cma.condition() << " "       // Column 2
        << cma.sigma() << " "           // Column 3
        << cma.solution().value << " ";     // Column 4
    copy(
         cma.solution().point.begin(),              
         cma.solution().point.end(),                // Column 5 & 6
         ostream_iterator< double >( results, " " ) 
         );
    copy( 
         cma.mean().begin(),                        // Column 7 & 8
         cma.mean().end(), 
         ostream_iterator< double >( results, " " ) 
          );
    results << endl;
 
    //plot << "plot 'results.txt' every ::2 using 7:8 with lp title 'Population mean'" << endl;
    plot << "plot 'results.txt' using 7:8 with lp title 'Population mean'" << endl;
 
    return( EXIT_SUCCESS ); 
}