MixtureGaussianEDAs_trajectory

 EXAMPLE 15:  Continuous EDAs that learn mixtures of distributions
              for  the trajectory problem (see previous examples for
              details on this problem)

0001 % EXAMPLE 15:  Continuous EDAs that learn mixtures of distributions
0002  %              for  the trajectory problem (see previous examples for
0003  %              details on this problem)
0004 
0005  global MGADSMproblem
0006  cd ../trajectory % The spacecraft-trajectory problem instance is in this directory
0007  load EdEdJ
0008  cd ../Mateda2.0
0009  
0010  NumbVar = 12;
0011  PopSize = 5000; 
0012  F = 'EvalSaga';
0013  Card(1,:) = [7000,0,0,0,50,300,0.01,0.01,1.05,8,-1*pi,-1*pi];
0014  Card(2,:) = [9100,7,1,1,2000,2000,0.90,0.90,7.00,500,pi,pi]; 
0015  cache  = [0,0,1,0,1]; 
0016   
0017  learning_params(1:5) = {'vars','ClusterPointsKmeans',10,'sqEuclidean',1};
0018  edaparams{1} = {'learning_method','LearnMixtureofFullGaussianModels',learning_params};
0019  edaparams{2} = {'sampling_method','SampleMixtureofFullGaussianModels',{PopSize,3}};
0020  edaparams{3} = {'replacement_method','best_elitism',{'fitness_ordering'}};
0021  selparams(1:2) = {0.1,'fitness_ordering'};
0022  edaparams{4} = {'selection_method','truncation_selection',selparams};
0023  edaparams{5} = {'repairing_method','SetWithinBounds_repairing',{}};
0024  edaparams{6} = {'stop_cond_method','max_gen',{5000}};
0025  [AllStat,Cache]=RunEDA(PopSize,NumbVar,F,Card,cache,edaparams)