Can I use the matrix to the variable genetic algorithm?

I want use matrix to variable.

 y=cost(gene)
  gene = ones(1,10)
    A = [0,find(gene(i,:))]; 
    Pre_Num = numel(A);
    S = 0;
    up = 0.6;
    Dec = 0;
   for j = 1:Pre_Num-1
     cost1 = exp(-(0.5*(A(j+1)-S)^2))
     term = dec_rel_a + (( A(j+1) -A(j))*(1 -up));
     Dec = A(j+1) - term;  
     predict = exp(-(0.5*(A(j+1)- Dec))^2));
    if (((1 - RM_a)*0.1) > (predict_rel_a - RM_a))
       dec_rel_a = 0;
       gene(i,A(j+1)=2;        
       temp_fit_val(j) = temp_fit_val(j) + 10000
       impro_rate = 1; 
    else
      temp_fit_val(j) = temp_fit_val(j) + 10                                  
      impro_rate = up;  
    end
    fit_val(i) = fit_val(i) + temp_fit_val(j);        
   Dec = Dec + (( A(j+1) - A(j))*(1 - impro_rate));
   S = A(j+1) - Dec; 
   end

I would like to know the optimal arrangement of the matrix using a genetic algorithm.

Answer 1

[Alan Weiss]

Sorry, I do not understand your code or your question. If you are using Global Optimization toolbox, then your individuals must be row vectors for the ga function. If you are trying to optimize individuals that are matrices, then convert each individual to a row vector using the command

xnew = x(:)';

To convert a row vector back to a matrix, use reshape.

Alan Weiss

MATLAB mathematical toolbox documentation

[Umber Saleem]

Alan Weiss can I use reshape for specifying constraints as well?