Matlab - input argument is not defined

Question

I have this very simple function in matlab.

function [f]=f1(a, xx)
xx
f = -exp(-a(1)*(xx(1)-1)^2 - a(2)*(xx(2)-1)^2) - exp(-a(1)*(xx(1)+1)^2 -a(2)*(xx(2)+1)^2);

It does print the value of xx, and then complains "Input argument "xx" is undefined". How is this possible? What's going on?

I call the function with feval Here is another example that doesn't use feval. I get the error Input argument "xxx" is undefined. Please help me, I have no idea what's going on and I'm stuck.

I have: function [ ans ] = f2( xxx ) xxx %f2 is 1/(1+xxx^2), the function for problem 2 ans = 1 / (1+xxx^2); end

one file is neville.m

Q = neville(x,xi,f2) %NEVILLE implements Neville method for polynomial interpolation

nplus1 = max(size(xi)); Q = zeros(nplus1); Q(:,1) =f2(xi);

for i = 2 : nplus1 for j = 2 : i Q(i, j) = ((x - xi(i-j)) * Q(i, j-1) - (x - x(i))*Q(i-1, j-1)) / (x(i) - x(i-j)); end; end;

The other file is neville_driver.m N = 6; aux = [0:6]; xi = -5 + 10*aux/N;

Q = neville(4, xi, f2)

how are you calling your function?

user564376
– user564376

2011-03-21 02:02:16 +00:00
Commented Mar 21, 2011 at 2:02 — user564376
– user564376, Commented Mar 21, 2011 at 2:02

user564376 · Accepted Answer · 2011-03-21 02:40:00Z

2

Looks like you're passing a function f2 into neville.m. Try using Q=neville(x,xi,@f2); Also, if xi is a vector, you should use element wise multiplication (and raising it to the nth power) using a dot (.) before the operation, else it will give an error. i.e., 1/(1+xxx.^2).

answered Mar 21, 2011 at 2:40

user564376

Sign up to request clarification or add additional context in comments.

4 Comments

Viktor Over a year ago

So, the idea is that in neville_driver.m I want to use f2 as fi, but in neville.m I want to have a function at argument. How exactly do I do this?

Viktor Over a year ago

fi is a function parameter. In this particular situation, I am passing f2 in the place of fi.

Viktor Over a year ago

I also get this error message:Attempt to execute SCRIPT neville as a function. What is going on???

user564376 Over a year ago

it means that you have not defined neville as a function. at the moment, it's just a plain matlab script, which you're probably trying to execute as a function either by passing variables or by using a function handle. this will result in an error. you should rewrite neville as a function.

usul · Accepted Answer · 2011-03-21 02:40:03Z

0

Is xx a function? Because the syntax xx(2) is a function call. Did you mean xx*2 ?

Edit: your first example code works fine for me:

feval(@f1,[2,2],[10,10]);

Prints xx and then the result.

Edit2: and the code for f1:

function [f]=f1(a, xx)
xx
f = -exp(-a(1)*(xx(1)-1)^2-a(2)*(xx(2)-1)^2)-exp(-a(1)*(xx(1)+1)^2-a(2)*(xx(2)+1)^2);
end

edited Mar 21, 2011 at 2:40

answered Mar 21, 2011 at 2:32

usul

8046 silver badges21 bronze badges

2 Comments

Viktor Over a year ago

Perhaps I should use handle when I use f2 as function argument?

usul Over a year ago

As far as I know, it's not necessary.

Collectives™ on Stack Overflow

Matlab - input argument is not defined

2 Answers 2

4 Comments

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

4 Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related