Why this piece of code gives error?

Question

Sadiq Akbar 2022-4-20

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1700825-why-this-piece-of-code-gives-error

评论： Steven Lord 2022-4-21

CostFunction='fun4sn0';
nPop=120;
nVar=4;%nVar=1000;
nPop_Initial=nPop;
nPop_Final=30;
MaxNFE=3e3;
MaxIter0=ceil(MaxNFE/((nPop_Initial+nPop_Final)/2));   % Approximate Maximum Iterations
Cr=0.5;
xmin=[0 0 0 0];
xmax=[10 10 pi pi];
%%%%%%%%%%%
NFE=0;
Gbest.Position=[];
Gbest.Cost=inf;
BestCosts=nan(1,MaxIter0);
nfe=BestCosts;
for i=1:nPop
    Velocity(i,:)=zeros(1,nVar); %#ok<*SAGROW>
    Position(i,:)=xmin+(xmax-xmin).*rand(1,nVar);
   Cost(i)=CostFunction(Position(i,:));
    PbestPosition(i,:)=Position(i,:);
    PbestCost(i)=Cost(i);
    PbestVel(i,:)=Velocity(i,:);
    
    if PbestCost(i)<Gbest.Cost
        Gbest.Position=PbestPosition(i,:);
        Gbest.Cost=PbestCost(i);
        Gbest.Velocity= PbestVel(i,:);
    end
end

2 个评论
显示无隐藏无

Jon 2022-4-20

what error does it give, please cut and paste entire error message

Sadiq Akbar 2022-4-20

Thank you very much dear Jon for your response. It gives the following error:

Subscript indices must either be real positive integers or logicals.

Error in WGA1 (line 37)

Cost(i)=CostFunction(Position(i,:));

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Steven Lord 2022-4-20

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1700825-why-this-piece-of-code-gives-error#answer_946740

在 MATLAB Online 中打开

Either make your CostFunction variable a function handle to the fun4sn0 function (which is the approach I'd prefer):

CostFunction= @fun4sn0;

or use feval to evaluate the function by name.

Cost(i) = feval(CostFunction, Position(i, :));

Both of these assume that your function returns a scalar, since you're assigning its output into one element of the Cost variable.

3 个评论
显示 1更早的评论隐藏 1更早的评论

Image Analyst 2022-4-21

在 MATLAB Online 中打开

@Steven Lord messing with function handles and feval() seems like a roundabout way to do it. Is that better than just calling the function directly like I suggested?

Cost(i) = fun4sn0(Position(i,:));

Steven Lord 2022-4-21

@Image Analyst That is another option, and is the easiest if you want to hard-code the cost function. I think that particular name put me in "this is likely going to be used in some form of optimization" mode and in that case (the "function function" scenario, see ode45, fminsearch, fzero, integral, etc.) to keep the code as a general purpose solver you wouldn't want to hard-code the function.

请先登录，再进行评论。