Assign Multiple variable error

I write to code about optimization algorithm. When I use code block as a;

function z=Sphere(x)

z=sum(x.^2);

The codes run succesfully But I want to use two variables x and y. instead of this, I write

function z=Sphere(x,y)

z=sum(x.^2,y.^2);

It giver error:

Not enough input arguments.

If you help about source of eror codes, I will appreciate. Thanks

function []=DEA()
    
    rng('default')
    
    N=10; 
    F=0.8; 
    CR=0.9; 
    
    max_iter=5; 
    
    D=5;
    low=-100; 
    high=100; 
    Fonksiyon='Sphere'; 
    
    function z=Sphere(x)
        z=sum(x.^2);
    end
    
    obj=zeros(1,N);
    mutant=zeros(N,D);
    
    for i=1:N
        for j=1:D
            pop(i,j)=low+rand()*(high-low);
        end
        obj(1,i)=eval(strcat(Fonksiyon,'(pop(i,:))'));
    end
    
    [minimum,min_indis]=min(obj);
    bestParams=pop(min_indis,:);
    mins=zeros(1,max_iter);
    sayac=0;
    iterasyon=1;
    while (iterasyon<=max_iter)
        for i=1:N
            
            n1=fix(rand*N)+1;
            while(i==n1)
                n1=fix(rand*N)+1;
            end
            n2=fix(rand*N)+1;
            while(i==n2||n1==n2)
                n2=fix(rand*N)+1;
            end
            n3=fix(rand*N)+1;
            while(i==n3||n1==n3||n2==n3)
                n3=fix(rand*N)+1;
            end
            
            for j=1:D
                mutant(i,j)=pop(n3,j)+(pop(n1,j)-pop(n2,j))*F;
                
                if mutant(i,j)>high
                    mutant(i,j)=high;
                end
                if mutant(i,j)<low
                    mutant(i,j)=low;
                end
            end
            
            trial=zeros(N,D);
            j=fix(rand*D)+1;
            for k=1:D
                if rand<CR || k==j
                    trial(i,k)=mutant(i,k);
                else
                    trial(i,k)=pop(i,k);
                end
            end
            
            score=eval(strcat(Fonksiyon,'(trial(i,:))'));
            
            if score<obj(1,i)
                obj(1,i)=score;
                pop(i,:)=trial(i,:);
            end
            
            if score<minimum
                minimum=score;
                bestParams=trial(i,:);
            end
            
        end
        
        fprintf('Iterasyon=%d .... En Küçük Değer=%g\n',iterasyon,minimum);
        mins(iterasyon)=minimum;    
        iterasyon=iterasyon+1;
    end
    end

[Mitch Lautigar]

Error Reasoning: MATLAB's sum function sums along a specific axis (either by rows, or by columns) for an array of data. The 2nd input you are giving it (y.^2) is being treated as dimensions for the sum function to sum along.

• source: https://www.mathworks.com/help/matlab/ref/sum.html

Solution: Change your sphere function as I show below. Doing this makes it more generic. The function doesn't care if the input variable for the function call is "a" or "x" and you can just call the function twice in your programming.

old:

function z=Sphere(x)

z=sum(x.^2);

end

new:

function z=Sphere(a)

z=sum(a.^2);

end

I also do hope you are calling the function "Sphere" somewhere in your code, because it currently isn't visible in the code you called. Since the "sphere" function is only accessible in the DEA code, MATLAB considers it a local function and will ONLY work inside the DEA Function

[Vatan Tosun]

I dont understan actually, my function is sum(x1^2+x2^2) so there is two variable. Your code consist only one variable 'a'. I want to sum two integer in my function.

[Mitch Lautigar]

Your original notation was:

z=sum(x.^2,y.^2); which was why I gave my original answer encouraging you to use the variable 'a' to avoid confusion.

New Notation:

sum(x1^2+x2^2) should be written as "sum(x1.^2+x2.^2)" and both "x1" and "x2" should be the same dimensions. If not it will error out.

[Vatan Tosun]

Ok I understand thank you. But I wonder How I write "x1" and "x2" in be the same dimensions to not give error.

[Mitch Lautigar]

"But I wonder How I write "x1" and "x2" in be the same dimensions to not give error."

x1 an x2 are inputs that you provide. You can either reshape the data (google MATLAB reshape) or you can add logic to sum the arrays regardless if they are the same size (see my code below).

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

%------------------------------------------------------------------------%

%Get simple math out of way

x1_1 = x1.^2;

x2_1 = x2.^2;

%------------------------------------------------------------------------%

%If statement to handle variable length x arrays

%if x1_1 is longer than x2_1

if length(x1_1) > length(x2_1)

x1_1(1:length(x2_1) = x1_1(1:length(x2_1))+x2_1;

output_x = sum(x1_1);

%if x2_1 is longer than x1_1

elseif length(x1_1) < length(x2_1)

x2_1(1:length(x1_1) = x2_1(1:length(x1_1))+x1_1;

output_x = sum(x2_1);

%arrays are same length

else

output_x = sum(x1_1 + x2_1);

end

%end suggested code

%------------------------------------------------------------------------%

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

This logic will add the values together and then sum them up regardless of the dimensions. I don't think their is a logic issue in my indexing, but you are welcome to try this food for thought and play around with it. If their is an issue, let me know.