plot the function value according to the value of the optimized parameter by genetic algorithm

Dears,

I would like to extract the value of the optimized parameters x* and the value of the function at each step of the optimization.

From my initial search on the internet there is a way using the plot functionality, but I was not able to get the funtion value Y in terms of the optmized parameters values X* at each step neither

The optimization methodology is GA.

Could you please help me with that.

[Asvin Kumar]

Does this help? https://www.mathworks.com/help/gads/examples/genetic-algorithm-options.html#gaoptionsdemo-3

[Zakaria]

Thank you for your response,

Unfortunatelly the following options does not show the optimized values at each iteration:

opts = optimoptions(@ga,'PlotFcn',{@gaplotbestf,@gaplotstopping});

[Asvin Kumar]

The options I mentioned in my previous comment plot the best and mean value of a particular generation at the specified Plot Interval.

I'm unsure what you're looking for. If you would like to extract the optimal parameters and optimal value at each generation, you can have a look at the following link or at Ameer Hamza's answer.

Display Options: https://www.mathworks.com/help/gads/genetic-algorithm-options.html#f17784

Depending on the nature of your variable of optimization, you can try using 'gaplotbestindiv' as mentioned at https://www.mathworks.com/help/gads/genetic-algorithm-options.html#f14474.

Let me know if this helps or if it doesn't, please clarify your question and I'll get back to you.

[Zakaria]

Thank you Asvin for the response,

Ameer's code works perfectly and answers my question.

Answer 1

[Ameer Hamza]

This example shows how to extract the best objective function value and the corresponding x vector at each iteration

global x_iterations y_iterations
x_iterations = [];
y_iterations = [];
obj_fun = @(x) sum(x.^2);
opts = optimoptions('ga', 'OutputFcn', @myOutFcn);
[x_final, f_final] = ga(obj_fun, 10, [], [], [], [], [], [], [], opts);
function [state, options, optchanged] = myOutFcn(options, state, ~)
    global x_iterations y_iterations
    [best_val, idx] = min(state.Score);
    best_x = state.Population(idx,:);
    x_iterations = [x_iterations; best_x];
    y_iterations = [y_iterations; best_val];
    optchanged = false;
end

[Zakaria]

Thank you Ameer,

This function indeed extracts the information I need for my analysis.

[Ameer Hamza]

Glad to be of help.