Recover the best individual of each population in GA

The documentation ga doc especifies that the output arguments consider the final population only. Is there a way to recover the best individual of each generation?

Answer 1

[Sam Chak]

Yes, @Jaime Trivino, it is possible according to this section.

state.Score     % gives you the scores of the current population
state.Best      % gives you the vector containing the best score in each generation

To do this, you must call the OutputFcn through specifying it the option

option = optimoptions('ga', 'OutputFcn', @gaoutfun)
[x, fval, exitflag, output] = ga(fun, nvars, A, b, Aeq, beq, lb, ub, nonlcon, options)

and then create the ga output function m-file that must have the following calling syntax:

function [state, options, optchanged] = gaoutfun(options, state, flag)
    % See example here:
    % https://www.mathworks.com/help/gads/custom-output-function-for-genetic-algorithm.html
end