Mean of columns within matrix blocks of different dimensions

I would like to create a matrix M where each column equals to the mean of the matrix blocks of the matrix A.

Case 1: The code below works well when I have subarrays of always 3 columns

B = reshape(A,2151,3,28);

M = squeeze(mean(B,2));

Case 2: But, now I need to group the columns of the matrix A either by 3 or 4.

B = mat2cell(A,2151,[3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 3 3 3]);

I created subarrays with mat2cell, but then I don't know how to average each blocks. The squeeze function doesn't work anymore.

Is it possible to produce a similar output than in Case 1 when I have subarrays of different dimensions ?

Thanks in advance for your help

Answer 1

[Sriram Tadavarty]

Hi Fpet,

You can try the following:

B = mat2cell(A,2151,[3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 3 3 3]);
bAvgCell = arrayfun(@(x) mean(B{x},2),1:length(B),'UniformOutput',false);
M = cell2mat(bAvgCell);

Hope this helps.

Regards,

Sriram

[fpet]

Wonderful, thanks