Add Cells/Matrices Row by Row

I have a large number of matrices/vectors. And I need to add then row by row. So I have 100 different 50x10 matrices. Now I want to add Row 1 (first cell) + Row 1 (second cell)+...., and repeat it for all 60 rows and get a final 50x10 matrix. I have them all in cell format.

[Image Analyst]

And your question is......?

My question is why do you have 100 different matrices? Hopefully they're not all separate variables but you're just getting a new matrix in a loop that's iterating 100 times. Or are all 100 matrices in individual cells of a 1-D cell array?

My second question is why are you insisting on cell arrays when, if the matrices are all the same size 50x10, when a normal 3-D double array would be much faster and more efficient?

Next question is : can you attach your data in a .mat file, or attach your script that generates the data?

Next question is did you just try a simple for loop - should be very very simple.

By the way, 100 is not large as far as array sizes or memory usage.

[Avik Mahata]

Thanks for taking time to reply. Below is the details. Any suggestions will be extremely helpful for my reserach.

Well here is the problem, I have a large matrix of about 600,000 rows and 14 coulumns. This data is a 6000 different bins with 100 columns each. So what I am trying to do is add (Row1 + Row101+ Row201....+..), (Row2 + Row 102+ Row 202 + ....) and get a final matrix of 100x14.

I am trying to elaborate more below,

a=rand(12,2)
a =
    0.5862    0.1559     % Row 1
    0.6509    0.7744     % Row 2
    0.8756    0.7171     % Rowl 3
    0.3699    0.7807      % Row 4
    0.5258    0.6832      % Row 5
    0.0857    0.9956      % Row 6
    0.0644    0.5571      % Row 7
    0.1754    0.6236       % Row 8
    0.6654    0.6843       % Row 9
    0.4061    0.1972       % Row 10
   0.7061    0.8972       % Row 11
   0.3061    0.9720       % Row 12

Now what I wanted to do is to bin 4 rows together. So we have 3 bins like below,

Bin 1
    0.5862    0.1559     % Row 1
    0.6509    0.7744     % Row 2
    0.8756    0.7171     % Rowl 3
    0.3699    0.7807      % Row 4
Bin 2
    0.5258    0.6832      % Row 5
    0.0857    0.9956      % Row 6
    0.0644    0.5571      % Row 7
    0.1754    0.6236       % Row 8
Bin 3 
    0.6654    0.6843       % Row 9
    0.4061    0.1972       % Row 10
   0.7061    0.8972       % Row 11
   0.3061    0.9720       % Row 12

Now what I wanted to do is take a average of the (Row1+ Row 5+ Row 9), (Row 2+ Row 6 + Row 10) and so on. Once I add them up I will end up with a matrix of , 4x 2.

What I trying is create a loop llike below hoping that the loop repeats 100 times and create 100 rows. But its not doing it somehow. I am doing something wrong in the loop.

clc
clear all
M=dlmread('segmentaf');
M(1:101:end,:) = [];
L = length(M);

for x = 1:1:100
    for a = 1:100:(L-101)
           for b = a+100
       Q = M(a:b,13);
         
        end 
    Q3 = Q1+Q2;
    end
    
end

Answer 1

[dpb]

"matrix of about 600,000 rows and 14 columns. This data is a 6000 different bins with 100 columns each. So what I am trying to do is add (Row1 + Row101+ Row201....+..), (Row2 + Row 102+ Row 202 + ....) and get a final matrix of 100x14. "

Using your smaller example, but same logic works, the "dead ahead" solution is

>> x=rand(12,2)         % some sample data
x =
    0.8147    0.9572
    0.9058    0.4854
    0.1270    0.8003
    0.9134    0.1419
    0.6324    0.4218
    0.0975    0.9157
    0.2785    0.7922
    0.5469    0.9595
    0.9575    0.6557
    0.9649    0.0357
    0.1576    0.8491
    0.9706    0.9340
>> 
% the engine
N=4;                    % number of elements in group
M=size(x,1)/N;          % number of bins -- must be evenly divisible
S=zeros(M,size(x,2);    % preallocate
for i=1:M
  S(i,:)=sum(x(i:N:end,:));
end

Show what we got...

>> S
S =
    2.4046    2.0347
    1.9682    1.4368
    0.5631    2.4416
>> 
>> x(1,1)+x(5,1)+x(9,1)     % illustrate first term gets right answer...
ans =
    2.4046
>>

[Avik Mahata]

Thanks for the response. But actually I am adding those 3 bins of 4x2 matrices to add row by row. So it should give a final 4x2 matrix.

[dpb]

So, switch N from 4 to 3 -- but that's not what your example asked...

"Now what I wanted to do is to bin 4 rows together. So we have 3 bins like below, "

The logic is generic regardless of the size as long as size(data,1)/N is integer.

>> N=3;                                     % changed mind, groups of 3, not four
>> M=size(x,1)/N;                           % compute # bins given N
>> S=zeros(M,2);                            % preallocate again
>> for i=1:M,S(i,:)=sum(x(i:N:end,:));end   % and compute
>> S                                        % what we get
S =
    2.9715    1.9270
    2.2426    2.7158
    2.1526    3.3057
    2.1568    0.9698
>> x(1,1)+x(4,1)+x(7,1)+x(10,1)             % again, just check got it right...
ans =
    2.9715
>>