Applying functions to each cell in a cell array

Good afternoon,

I have a 60 by 1 cell array that stores matrices of different sizes (all numeric). I want to perform certain functions (mean, diff, hypot, convhull...etc) on these matrices from the 1st element of the cell array, to the 60th element.

Is there a way I can do that?

Thank you.

[Benedict Low]

Thank you all for your responses. So suppose I have the following array, consisting of x and y coordinates of various points

c1=[1,2; 3,4; 5,6]; % 3 x-coordinates and 3 y-coordinates
c2=[2,10; 4,8; 9,3; 10,2; 4,5]; % 5 x and y coordinates
c3=[3,8; 9,1; 7,10; 5,1]; % 4 x and y coordinates
C={c1; c2; c3};

is there a way I can calculate the area of the convhull for each cell of the array?

[Jan]

Did you read the answers, which have been given already?

[Benedict Low]

Yes I did Jan. But ran into errors. Trying to fix them now.

Answer 1

[the cyclist]

cellfun does exactly this.

Answer 2

[Moses Wayne]

The function call "cellfun" is a useful function in this case. In the case of the function mean, the code below would work to get the mean of each subset within the cell array:

myCell = {[0 1 2]; [3 4 5]; [6 7 8]; [9 10 11]}
result = cellfun(@mean,myCell)

In this case, result would be an array of values [1 4 7 10]. I've linked the documentation to "cellfun" here.

Answer 3

[Adam]

If you have multiple functions to perform a for loop would likely be simplest. The idea that for loops are very slow in Matlab is often erroneous. I favour cellfun and arrayfun from an aesthetic perspective, but from a speed perspective for loops are usually faster. And certainly if you want to calculate multiple things then a single pass over the data in a for loop is a lot better than multiple calls to cellfun, though you could of course create a composite function that will calculate all your results on a single cell and then call this via cellfun if you prefer the cellfun approach.