Extract first and last nonzero elemenents inside subarrays avoiding mat2cell

Can somone please give me a hand?

I have a matrix A such as:

A = [0 0 1 2 0 5 0 0 6 0 3 0 ; 0 3 0 2 6 0 0 0 0 0 0 0];

I'd like to check where are the first and last nonzero values along my subarrays of 6 elements on each row. So the expected Output shoul be:

Out = [0 0 1 0 0 1 0 0 1 0 1 0 ; 0 1 0 0 1 0 0 0 0 0 0 0];

I know I can easilly use mat2cell to perform this but I'd like to avoid using this funciton.

Thank's for the help,

Santos

Answer 1

[Bruno Luong]

A = [0 0 1 2 0 5 0 0 6 0 3 0 ; 0 3 0 2 6 0 0 0 0 0 0 0];

b = reshape(A.',6,[]) ~= 0;
[m,n] = size(b);
[~,istart] = max(b,[],1);
[~,istop] = max(flip(b,1),[],1);
i = [istart; m+1-istop] + (0:n-1)*m;
B = zeros(size(b));
B(i) = b(i);
Out = reshape(B,flip(size(A))).'

[Santos García Rosado]

Thank you Bruno!

Answer 2

[DGM]

I arbitrarily chose to avoid loops and find() and ended up with this ugly thing. It works, but I wouldn't call it elegant. I added some dummy values to the test array to demonstrate that it handles cases where there are no matches

A = [0 0 1 2 0 5 0 0 6 0 3 0 ; 0 0 0 0 0 0 0 0 0 0 0 0; 0 3 0 2 6 0 0 0 0 0 0 0];
A=reshape(A',6,[])'

matches=[sum(cumprod(A == 0,2),2)+1; 6-sum(cumprod(A == 0,2,'reverse'),2)];
matches=[[1:6 1:6]; matches']';
matches=matches(matches(:,2)>0 & matches(:,2)<7,:);

B=zeros(size(A));
B(sub2ind(size(A),matches(:,1),matches(:,2)))=1;
B=reshape(B',12,[])'

this yields:

B =
     0     0     1     0     0     1     0     0     1     0     1     0
     0     0     0     0     0     0     0     0     0     0     0     0
     0     1     0     0     1     0     0     0     0     0     0     0

Again, I added the extra row

[Santos García Rosado]

Thank you DGM. Check Bruno's answer. You might like the code he came up with.