using fzero with arrayfun searching for zeros inside an interval

I have a function f = @(x) -x.^2+4 with a zero at -2 and +2

using fzero(f, [-3, 1.9]) I get the (correct) zero inside the interval [-3, 1.9] , which is -2

when I now have not one but multiple intervals where I want to detect zeros, let's say: [-3, 1.9] and [1.9, +3],

fzero(f, [-3, 1.9; 1.95, 3]) returns an error, telling me, that the second argument must be scalar or a vector of length 2 (= interval).

When I now try to use arrayfun in order to successively apply my intervals to fzero I get:

arrayfun(@(x) feval("fzero" , f, x), [-3 1.9; 1.95 3])

ans =

-2 2

2 2

because fzero does not take intervals, but single values.

However I would like to use intervals and the expected result should be

-2

2

I do not want to use a for loop because of performance issues.

Now the question: Can anyone help how to make "fzero" run on multiple intervals using "arrayfun" (or any other trick) ?

Answer 1

[Ameer Hamza]

Create a cell array and then use cellfun()

f = @(x) -x.^2+4;
C = {[-3 1.9]; [1.95 3]};
sol = cellfun(@(x) fzero(f, x), C) 

[Thomas]

Thank you very much - this works fine!