What is wrong with the function code when it can work perfectly without function code

Hi There!

I am having trouble on my code where I made the code in function [input] = f(output) doesn't give me the expected answer but a normal script do.

function [k,p,err,P] = fixpt(g,po,tol,maxi)
% g is the function input as function handle
% po is the first guess
% tol is the tolerance
% maxi is the maximum iteration
% K is the number of iteration
% p is the sequence {pn}
% error is the absolute error
P(1) = po
    for k = 2:maxi
        P(k) = g(P(k-1));
        err = abs(P(k) - P(k-1));
        relerr = (err/abs(P(k)));
        p = P(k);
        if abs(err)<tol || abs(relerr)<tol
            return
        end
    end
    if k == maxi
        fprintf('Iteration exceed maximum.')
    end
end

If I do run it:

fixpt(@(x) sin(x)-1,-1,0.001,100)

It doesn't give me the the answer I expected whereas:

g = @(x) sin(x)-1
maxi = 100
tol = 0.00001
po = -1
P(1) = po
for k = 2:maxi
        P(k) = g(P(k-1));
        err = abs(P(k) - P(k-1));
        relerr = (err/abs(P(k)));
        p = P(k);
        if abs(err)<tol || abs(relerr)<tol
            return
        end
    end
if k == maxi
    fprintf('Iteration exceed maximum.')
end

This does.

Im using R2021a. Thank you.

Answer 1

[Torsten]

Works for me.

[k,p,err,P] = fixpt(@(x) sin(x)-1,-1,0.00001,100)

k = 12

p = -1.9346

err = 1.0380e-05

P = 1×12

-1.0000 -1.8415 -1.9636 -1.9238 -1.9383 -1.9332 -1.9350 -1.9344 -1.9346 -1.9345 -1.9346 -1.9346

function [k,p,err,P] = fixpt(g,po,tol,maxi)
% g is the function input as function handle
% po is the first guess
% tol is the tolerance
% maxi is the maximum iteration
% K is the number of iteration
% p is the sequence {pn}
% error is the absolute error
P(1) = po;
for k = 2:maxi
    P(k) = g(P(k-1));
    err = abs(P(k) - P(k-1));
    relerr = (err/abs(P(k)));
    p = P(k);
    if abs(err)<tol || abs(relerr)<tol
        return
    end
end
if k == maxi
    fprintf('Iteration exceed maximum.')
end
end

[Kenichi]

I don't get it, when i run it on my MATLAB it stopped on the second iteration

[Kenichi]

It gives this error message:

Output argument "p" (and maybe others) not assigned during call to "fixpt".

[Torsten]

Maybe you used a different value for "tol" in both cases ?

[Kenichi]

I'm not sure if that's the case, since different value of "tol" would still give the array of values..

[Kenichi]

It seems like my indentation is wrong, I have seen your code and fixed it. Thank you!

[Torsten]

Maybe you should restart MATLAB.

Answer 2

[Davide Masiello]

The value of for tol differs of two orders of magnitude in the examples you gave (i.e., you used 0.001 in the function call, 0.00001 in the functionless script).

[k,p,err,P] = fixpt(@(x) sin(x)-1,-1,0.00001,100)

P = -1

k = 12

p = -1.9346

err = 1.0380e-05

P = 1×12

-1.0000 -1.8415 -1.9636 -1.9238 -1.9383 -1.9332 -1.9350 -1.9344 -1.9346 -1.9345 -1.9346 -1.9346

function [k,p,err,P] = fixpt(g,po,tol,maxi)
% g is the function input as function handle
% po is the first guess
% tol is the tolerance
% maxi is the maximum iteration
% K is the number of iteration
% p is the sequence {pn}
% error is the absolute error
P(1) = po
    for k = 2:maxi
        P(k) = g(P(k-1));
        err = abs(P(k) - P(k-1));
        relerr = (err/abs(P(k)));
        p = P(k);
        if abs(err)<tol || abs(relerr)<tol
            return
        end
    end
    if k == maxi
        fprintf('Iteration exceed maximum.')
    end
end

Is this correct?

[Kenichi]

No, what I expected is a list that contains the iterations. In this code, when I it in the form of function, i can't get those values.

[Davide Masiello]

If you copy-paste the code above in a new script, does it work?

[Kenichi]

Yeah it does work, it seems like my indentation is the problem. Sorry that I could only choose one answer. Anyways, Thank you and have a great day.