Why the given codes give errors for vector bounds but runs well for scalar bounds?

Question

Sadiq Akbar 2023-7-31

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2002557-why-the-given-codes-give-errors-for-vector-bounds-but-runs-well-for-scalar-bounds

评论： Sadiq Akbar 2023-8-1

I have downloaded an algorithm BBO from the Mathworks site. I run it with my fitness function "fitVectorized1.m". I oberved the following;

1- When I run it for scalar bounds, it works and gives reults

But

2- When I run it for vector bounds, it gives me errors.

What to do?

4 个评论
显示 2更早的评论隐藏 2更早的评论

KSSV 2023-7-31

在 MATLAB Online 中打开

u=[1 5 30 70];dim=length(u);
[BestX,fmin]=bbo3(@(b)fitVectorized1(b,u),dim,0,90,100,50)
Unrecognized function or variable 'RouletteWheelSelection'.
Error in bbo3 (line 75)
                j=RouletteWheelSelection(EP);

Sadiq Akbar 2023-7-31

RouletteWheelSelection.m

Sorry for forgetting to upload the rrquired file. Here it is attached.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

KSSV 2023-7-31

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2002557-why-the-given-codes-give-errors-for-vector-bounds-but-runs-well-for-scalar-bounds#answer_1280987

在 MATLAB Online 中打开

Use this function instead of your bbo3.

function [BestX,fmin]=bbo3(obj,dim,lb,ub,iter,pop1)
%% Problem Definition
CostFunction=obj; % Cost Function
nVar=dim;  % Number of Decision Variables
VarSize=[1 nVar];   % Decision Variables Matrix Size
VarMin=lb;   % Decision Variables Lower Bound
VarMax=ub;   % Decision Variables Upper Bound
%% BBO Parameters
MaxIt=iter;  % Maximum Number of Iterations
nPop=pop1;%50;            % Number of Habitats (Population Size)
KeepRate=0.01;%0.2;                   % Keep Rate
nKeep=round(KeepRate*nPop);     % Number of Kept Habitats
nNew=nPop-nKeep;                % Number of New Habitats
% Migration Rates
mu=linspace(1,0,nPop);          % Emmigration Rates
lambda=1-mu;                    % Immigration Rates
alpha=0.9;
pMutation=0.1;
sigma=0.02*(VarMax-VarMin);
%% Initialization
% Empty Habitat
habitat.Position=zeros([],dim);
habitat.Cost=[];
% Create Habitats Array
pop=repmat(habitat,nPop,1);
% Initialize Habitats
for i=1:nPop
    pop(i).Position=unifrnd(VarMin,VarMax,VarSize);
    pop(i).Cost=CostFunction(pop(i).Position);
 
end
% Sort Population
[~, SortOrder]=sort([pop.Cost]);
pop=pop(SortOrder);
% Best Solution Ever Found
BestSol=pop(1);
% Array to Hold Best Costs
BestCost=zeros(MaxIt,1);
%% BBO Main Loop
for it=1:MaxIt
    
    newpop=pop;
    for i=1:nPop
        for k=1:nVar
            % Migration
            if rand<=lambda(i)
                % Emmigration Probabilities
                EP=mu;
                EP(i)=0;
                EP=EP/sum(EP);
                
                % Select Source Habitat
                j=RouletteWheelSelection(EP);
                
                % Migration
                newpop(i).Position(k)=pop(i).Position(k) ...
                    +alpha*(pop(j).Position(k)-pop(i).Position(k));
                
            end
            
            % Mutation
            if rand<=pMutation
                newpop(i).Position(k)=newpop(i).Position(k)+sigma(k)*randn;  %<------ Changed here 
            end
        end
        
        % Apply Lower and Upper Bound Limits
        newpop(i).Position = max(newpop(i).Position, VarMin);
        newpop(i).Position = min(newpop(i).Position, VarMax);
        
        % Evaluation
        newpop(i).Cost=CostFunction(newpop(i).Position);
    end
    
    % Sort New Population
    [~, SortOrder]=sort([newpop.Cost]);
    newpop=newpop(SortOrder);
    
    % Select Next Iteration Population
    pop=[pop(1:nKeep)
         newpop(1:nNew)];
     
    % Sort Population
    [~, SortOrder]=sort([pop.Cost]);
    pop=pop(SortOrder);
    
    % Update Best Solution Ever Found
    BestSol=pop(1);
    
    % Store Best Cost Ever Found
    BestCost(it)=BestSol.Cost;
    
    % Show Iteration Information
    %disp(['Iteration ' num2str(it) ': Best Cost = ' num2str(BestCost(it))]);
    
end
%% Results
% % figure;
% % %plot(BestCost,'LineWidth',2);
% % semilogy(BestCost,'LineWidth',2);
% % xlabel('Iteration');
% % ylabel('Best Cost');
% % grid on;
BestX=BestSol.Position % By Me
fmin=BestSol.Cost % By Me