%
% This calls a pareto genetic algorithm routine which is used to perform 
% multidimensional objective optimization (MOO) in the case where there are 
% *two* continious functions to minimize
% 
% See EPage 117 of Haupt and Haupt for the example considered here
%
% Written by:
% -- 
% John L. Weatherwax                2006-08-28
% 
% email: wax@alum.mit.edu
% 
% Please send comments and especially bug reports to the
% above email address.
% 
%-----

clear; 
%close all; 
clc; 

randn('seed',0); rand('seed',0);

addpath('../Chapter3');

% Parameters used in the genetic algorithm
% 
popsize   = 300;  % the population size
selection = 0.5;  % selection rate
mutrate   = 0.02; % the mutation rate
maxit     = 50; 

% a function to minimize (and its bounds)
%
bounds = [ 1, 2 ; 1, 2; 1, 2; 1, 2 ]; % the true functions domain 

% our function needs to be able to evaluate an entire population and produce two columns
% for the evaluate of f_1 and f_2
%
wfn = @(x) chap_5_prob_2_original_fns(x) ; 

[pop,pop_costs] = pareto_GA( wfn, bounds, popsize, selection, mutrate, maxit );

disp('top 5 solution x values');
pop(1:5,:)
disp('top 5 f(x) values');
pop_costs(1:5,:)

%saveas(gcf,'../../WriteUp/Graphics/Chapter5/chap_5_prob_3_pareto_front.eps','epsc');