Error using arrayfun : Non scalar in Uniform output

Hello,

I have a cell array (Z) of size (480*1) where in each row I have the individual double arrays that I want(e.g. 1*70 double). Now I want to apply a custom function to each row (each elements of the cell), for this I use arrayfun:

C = arrayfun(@ (x) (((a/length(Z))*(1/(2*pi))).*((abs(fftshift(fft(Z,1024))).^2))),1:480);

and I get this error:

Error using arrayfun Non-scalar in Uniform output, at index 1, output 1. Set 'UniformOutput' to false.

(Worth mentioning that I zeropad my fft, so the output data are all the same size(1*1024))

Answer 1

[Azzi Abdelmalek]

Add 'uniformoutput',false or 'un',0

   C = arrayfun(@ (x) (((a/length(Z))*(1/(2*pi))).*((abs(fftshift(fft(Z,1024))).^2))),1:480,'un',0);

[Mona Mahboob Kanafi]

Thanks for your answer, but it shouldn't be the problem of output size since all outputs do to use of zeropadding must be of size 1*1024. With your correction the error will be:

Undefined function 'fft' for input arguments of type 'cell'.

Error in @(x)(((a/length(Z))*(1/(2*pi))).*((abs(fftshift(fft(Z,1024))).^2)))

[Azzi Abdelmalek]

Change fft(Z,1024) by fft(x,1024)

C = arrayfun(@ (x) (((a/length(Z))*(1/(2*pi))).*((abs(fftshift(fft(x,1024))).^2))),1:480,'un',0);

[Mona Mahboob Kanafi]

x in your expression is 1:480, while fft is going to be applied on each element of Z(as explained in my question) i.e. Z{1} which has 1*70 elements , then Z{2} having 1*58 elements,...

[Azzi Abdelmalek]

Ok, then use cellfun

C = cellfun(@ (x) (((a/length(Z))*(1/(2*pi))).*((abs(fftshift(fft(x,1024))).^2))),Z,'uniformoutput',false);