These are 2 subprograms i have used to define 2 function grad and fnc. I have called the function grad , and in the function grad fnc is called, now x is a vector, and fnc is a scalar function of a vector variable still it is returning a vector

the second thing is that an error '??? Subscript indices must either be real positive integers or logicals' is coming... the error comes in the line 'grad1(k,1)=(fnc(x1)-fn(x))/(0.001)' of the grad func subprogram.

function N5=grad(fnc,x)

grad1=zeros(length(x),1);

x1=x;

for k=1:1:length(x)

     x1(k,1)=x1(k,1)+0.001;

    grad1(k,1)=(fnc(x1)-fn(x))/(0.001);

   x1=x;

   end

N5=grad1;

function N4= fnc(x)
N4=(x(1,1)*x(1,1)+x(2,1)-11.0)^2+(x(1,1)+x(2,1)*x(2,1)-7)^2;

Answer 1

[Marta Salas]

The problem is that fnc is an input for grad, so fnc is a variable from now on. When you try to call fnc(x), MATLAB thinks you're trying to access position x on the array fnc and returns an error because x is not an integer.

I think what you want is:

function N5=grad(x)
grad1=zeros(length(x),1);
x1=x;

for k=1:length(x)
   x1(k,1)=x1(k,1)+0.001;
   grad1(k,1)=(fnc(x1)-fnc(x))/(0.001);
   x1=x;
end
N5=grad1;

function N4 = fnc(x) 
  N4=(x(1,1)*x(1,1)+x(2,1)-11.0)^2+(x(1,1)+x(2,1)*x(2,1)-7)^2;

[Arpit]

thank you

Answer 2

[Walter Roberson]

When you call grad() pass the function handle of fnc as the first argument

 grad(@fnc, rand(1,50))      %for example