Genetic Algorithm in Matlab

Genetic Algorithm in Matlab : Basic Implementation .

 %%steps
  %1.Initialize the populations 
  tic
  clc; clear all; close all;
  par_size=6;
  %Encode
  for i=1:par_size
      par(i,:)=dec2bin(round(1+1023*rand),10);
      a(i) = bin2dec(par(i,:));
  end

for iter = 1:100
%2.Evaluate fitness
fitness = zeros(1,par_size);
%The fitness function
%count the no of one's in the string
fitness = [sum(par=='1',2)].';
sum_fitness = sum(fitness)
prob_fitness = fitness./sum_fitness;
[~,I]=sort(prob_fitness);
%I is the order of the fitness value
prob_fitness_sort = prob_fitness(I);
%sorting
par = par(I,:);
%3.Select Parents
%Roulette wheel selection
%generate cumulative probability
cum_prob = zeros(size(prob_fitness));
A = zeros(size(prob_fitness));
for i=1:par_size
    A(i) = prob_fitness_sort(i);
    cum_prob(i) = sum(A);  
end
new_pop = par;
%Selection being done
r1 = zeros(1,par_size);
for i=1:par_size
    r1(i) = rand;
    if r1(i)<=cum_prob(1)
        new_pop(i,:) = par(1,:);
    elseif r1(i)>cum_prob(1) && r1(i)<=cum_prob(2)
        new_pop(i,:) = par(2,:);
    elseif r1(i)>cum_prob(2) && r1(i)<=cum_prob(3)
        new_pop(i,:) = par(3,:);
    elseif r1(i)>cum_prob(3) && r1(i)<=cum_prob(4)
        new_pop(i,:) = par(4,:);
    elseif r1(i)>cum_prob(4) && r1(i)<=cum_prob(5)
        new_pop(i,:) = par(5,:);
    elseif r1(i)>cum_prob(5) && r1(i)<=cum_prob(6)
        new_pop(i,:) = par(6,:);
    end
end
%new_pop is the new particle sample
%4.Crossover and Mutate

%crossover
pc = 0.6;   %pc is the crossover probability
for i=1:2:par_size %selecting adjacent couples
    r2 = rand;
    if r2<=pc
        l = size(new_pop, 2);
        %choose the parents
        %here adjacent parents are taken
        breeders(1,:)=new_pop(i,:);
        breeders(2,:)=new_pop(i+1,:);
        %choose a crossover point
        cp = randperm(l, 1);
        [i,cp];
        %do crossover
        b1 = [breeders(1, 1:cp), breeders(2, cp+1:end)];
        b2 = [breeders(2, 1:cp), breeders(1, cp+1:end)];
        new_pop(i,:) = b1;
        new_pop(i+1,:) = b2;
    end
end
%mutation
%pm be mutation probability
pm = 0.1;
for i=1:par_size
  r2 = rand;
  if r2<=pm %mutation probability is 0.1
      %Do mutation
      %select random position
      r3 = 1+round((10-1)*rand);
      [i,r3];
      new_pop(i,r3)=num2str(abs(str2num(new_pop(i,r3))-1));
  end
end
par = new_pop;
end

Is it correct ? Can it be improved ??