How to write selection function in genetic algorithm(Global Optimization Toolbox: ga)

Hi guys.

I have some problem to use the genetic algorithm.

I want to use customized selection function, but I can only see error message: Index exceeds matrix dimensions

Here is my code below:

function parents=AGAselection(expectation, nParents, options)

% nParents= the number of parents
% A= sorted expectation
% B= expectation of nParents top parents
% GBparents= global best parents
% 


expectation=(expectation(:,1));
parents=zeros(1,nParents);

%% Sorting accordance with fitness value
A=sort(expectation);
B=A(1);

%% Select Global Best
GBparents(1,1)=find(expectation==B(1,1));

%% Select Parents Randomly
RN=randperm(nParents);
parents=[GBparents RN(1:nParents-1)];

end

Please check my code and tell me what is the problem.

Thank you.

Answer 1

[Alan Weiss]

I am not sure, but I am suspicious of the line

GBparents(1,1)=find(expectation==B(1,1));

What makes you think that expectation==B(1,1) has just one element?

Alan Weiss

MATLAB mathematical toolbox documentation

[Jungeon Kim]

https://www.mathworks.com/matlabcentral/answers/195919-writing-a-custom-selection-function-for-a-genetic-algorithm-problem

The code is referred to the page above.

He said his problem is solved.

I don't know the structure of parents(I think the parents is vector that consists of cell.)

I want to extract just best parent.

Is there any problem in the line?