How can I reshape single column vector list into equally sized matrix columns

Hi,

I have a single column vector variable with a list of 26599 elements. I want to reshape this list into a matrix of equally sized columns of 512. The issue is that the number 26599 isn't divisible by 512 and so I get an error. Hence, I want to reshape the vector into matrix colums where all columns have the size of 512 except for the last one.

How would I go about doing this?

I tried the reshape function but that didnt work because of the issue mentioned above.

>> R = reshape(sample_test,512,[]);
Error using reshape
Product of known dimensions, 512, not divisible into total number of elements, 26599.

Thank you!

Answer 1

[Voss]

You can't have a matrix with different size columns, but you could fill the last column with NaNs:

sample_test = ones(520,1);

% append NaNs to sample_test up to the next multiple of block_size
block_size = 512;
N = block_size*ceil(numel(sample_test)/block_size);
sample_test(end+1:N) = NaN;

R = reshape(sample_test,512,[])

R = 512×2

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 NaN 1 NaN

[lil brain]

If I run your code underneath the comment I get an error:

Attempt to grow array along ambiguous dimension.

What am I doing wrong?

[Voss]

Sounds like sample_test is a matrix rather than a column vector. Check on that.

[lil brain]

Yes it is! My bad. I am fairly new to this. How do I proceed in this case?

[Voss]

You can make sample_test a column vector like this:

sample_test = sample_test(:);

and then use my code on that.

[lil brain]

Beautiful! This works!

Answer 2

[Torsten]

s = mod(26599,512);
n = 26599 - s;
m = n/512;
R = horzcat(reshape(sample_test(1:n),512,m),vertcat(sample_test(n+1:end),NaN(512-s,1)));

[lil brain]

When running this code I get the error:

Index exceeds the number of array elements (1040).

What am I doing wrong?

[Torsten]

Corrected, I think.