Getting Complex number from a function

function fxy = fxy(x,y)
    square = sqrt(x*x + y*y);
    t1 = (sin(20*square))/(20*square);
    t2 = (1/5)*cos(10*square);
    t3 = (y/2) - .3;
    fxy = t1 + t2 + t3;
    return
end

I am using the values below to run through the function above as a part of a machine learning project, but I keep recieving the error "Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 2.368008e-33" when I click run. The numbers I recieve for fxy are complex numbers, but if you plug the values in individually, there is no way for them to come out as complex. I am not sure how to get the correct results. (The function code is saved as fxy.m and run in the same folder).

dim = 100;
%create 100x100 array
a = linspace(-1, 1, dim);
b = linspace(-1, 1, dim);
[ax, by] = meshgrid(a,b);
%plug values into f(x,y)
fxy = fxy(ax, by);

Answer 1

[Sulaymon Eshkabilov]

Here is the corrected code:

dim = 100;
%create 100x100 array
a = linspace(-1, 1, dim);
b = linspace(-1, 1, dim);
[ax, by] = meshgrid(a,b);
%plug values into f(x,y)
fxy = FXY_FUN(ax, by);
meshc(fxy)

function fxy = FXY_FUN(x,y)                 % Be careful with the function name and variable name  
    square = sqrt(x.^2 + y.^2);             % Elementwise operation is necessary
    t1 = (sin(20*square))./(20*square);     % Another elementwise operation
    t2 = (1/5)*cos(10*square);
    t3 = (y/2) - .3;
    fxy = t1 + t2 + t3;
    return
end

[Eyan Hudson]

Thank you! I completely forgot about the elementwise operator, that fixed it!

[Sulaymon Eshkabilov]

Most welcome!