Split a row into several rows

29 views (last 30 days)
Olu adroit on 14 Jan 2016
Commented: Olu adroit on 14 Jan 2016
Hi all, I am trying to split a single-row matrix say for example
A = [1,2,3,4,17,18,19,20,33,34,35,36,49,50,51,52,65,66]
into a multiple-row matrix B, the maximum number of columns in B being 5. I would expect to get something like
B = [1,2,3,4,17;18,19,20,33,34;35,36,49,50,51;52,65,66]
I tried using the reshape function as below but it's giving an error saying
Product of known dimensions, 5, not divisible into total number of elements, 18
Please any suggestion would be appreciated.
%
% I would like to get
% B = 1 2 3 4 17
% 18 19 20 33 34
% 35 36 49 50 51
% 52 65 66
%
% code
A = [1,2,3,4,17,18,19,20,33,34,35,36,49,50,51,52,65,66];
B = reshape(A, 5, []);

Stephen23 on 14 Jan 2016
Edited: Stephen23 on 14 Jan 2016
A matrix cannot have gaps or missing values: every row must be the same length, just as every column must too. So you will have to join some values onto A before reshaping it:
>> N = 5;
>> reshape([A,nan(1,N-mod(numel(A),N))],[],N)
ans =
1 17 33 49 65
2 18 34 50 66
3 19 35 51 NaN
4 20 36 52 NaN
Olu adroit on 14 Jan 2016
Thanks for this Steve. What if I want to pad with 0 and not NaN? I subsequently exported the multi-row matrix as a csv file into Abaqus software, which is giving me error message because of the exported NaN. Thanks in advance.
Olu adroit on 14 Jan 2016
ok i got it now. I substituted NaN with zeros. Thanks

Ilham Hardy on 14 Jan 2016
The error message indicates the error very clear. What will be the dimension of the B matrix then? 5x4.9 matrix?
Unless you pad the A matrix beforehand with
A = [A,NaN,NaN];
the reshape command will not works (obviously).