How can I store the value at each iteration of a genetic algorithm?

I have a simple function and two constraints for my genetic algorithm code

ObjFcn = @myFitness;
nvars = 2;   %number of variables
LB = [0 0];  %lower bound
UB = [1 13]; %upper bound

ConsFcn = @myConstraints;
[x,fval] = ga(ObjFcn,nvars,[],[],[],[],LB,UB,ConsFcn,opts);

function y = myFitness(x)
y = 100*(x(1)^2-x(2))^2 + (1-x(1))^2;
end

function [c,c_eq] = myConstraints(x)
c = [x(1)*x(2) + x(1) - x(2) + 1.5; 10 - x(1)*x(2)];
c_eq = [];
end

I would like to store the values of x at each function evaluation, as well as keep track of the generation. I've found a similar question here but I don't really understand what the solution meant as it wasn't very clear. I believe I need to tell my fitness function to store values of the GA function so when the GA function reads the fitness function the input parameters at that iteration are saved... but I would like help on how the syntax would work for such a thing. Any help is appreciated!

I've also looked at the GA options and was unable to find help with my specific issue.

Answer 1

[Stephan]

Hi,

you coultd try this way:

[x,fval,vals] = solve_problem 

function [x,fval,vals] = solve_problem 
ObjFcn = @myFitness;
nvars = 2;   %number of variables
LB = [0 0];  %lower bound
UB = [1 13]; %upper bound
ConsFcn = @myConstraints;
iter = 1;
vals = [];

[x,fval] = ga(ObjFcn,nvars,[],[],[],[],LB,UB,ConsFcn);
function y = myFitness(x)
y = 100*(x(1)^2-x(2))^2 + (1-x(1))^2;
vals(iter,1:2) = x;
iter = iter+1;
end

function [c,c_eq] = myConstraints(x)
c = [x(1)*x(2) + x(1) - x(2) + 1.5; 10 - x(1)*x(2)];
c_eq = [];
end
end

After running this you can plot how x(1) and x(2) change over the iterations:

yyaxis left
plot(vals(:,1))
yyaxis right
plot(vals(:,2))

Best regards

Stephan

[Sylvain Cornelus]

Thank you Stephan! Exactly what I was looking for. I'll read up on the ga function to figure out how to store the generation.

Answer 2

[UNAL]

You can also extract the data directly from the graph that is obtained by configuring the options:

options = optimoptions('ga','PlotFcn', @gaplotbestf);

After obtaining the graph you can extract ist data:

fig = gcf;
dataObjs = findobj(fig,'-property','YData');
y1 = dataObjs(1).YData;
x1 = dataObjs(1).XData;
figure(7)
plot(x1,y1,'*r')

[djamel]

Many thanks for your feedback , i was seraching on this solutions for long time .

It was very useful for me .

Many thanks .