Efficiently converting a 3d matrix to a 2d matrix

973 visualizzazioni (ultimi 30 giorni)
Hello,
I would like to convert a 3d matrix into a 2d matrix. I want the 3rd dimension to be concatenated along dimension 1 in the 2d matrix. In the code below, the variable 'desired' illustrates what I want to achieve, but I want to do it more efficiently than via a for a loop. I tried doing the same with reshape (see variable B) but can't get it to produce the output I desire.
A(:,:,1) = ones(2,2);
A(:,:,2) = 2*ones(2,2);
A(:,:,3) = 3*ones(2,2);
desired = [];
for n = 1:size(A,3)
desired = cat(1,desired,A(:,:,n));
end
B = reshape(A,[],size(A,2),1);
Any advice would be much appreciated. Thanks for reading.

Risposta accettata

the cyclist
the cyclist il 21 Mar 2011
You can permute the 2nd and 3rd dimension before doing the reshape:
>> C = permute(A,[1 3 2]);
>> C = reshape(C,[],size(A,2),1)
  9 Commenti
Sem Diaz
Sem Diaz il 4 Dic 2019
How can you convert it back?

Accedi per commentare.

Più risposte (2)

Andrew Newell
Andrew Newell il 21 Mar 2011
You can do it with cell arrays:
Acell = num2cell(A,[1 2]); % put each page of A into a cell
Acell = reshape(Acell,size(A,3),1); %make the cell array a vector
desired = cell2mat(Acell);
EDIT: the solution by @the cyclist is nicer for this particular problem, but be careful about generalizing it. For example, suppose you want to pack your matrices into a 2x2 array:
A(:,:,1) = ones(2,2);
A(:,:,2) = 2*ones(2,2);
A(:,:,3) = 3*ones(2,2);
A(:,:,4) = 4*ones(2,2);
Acell = num2cell(A,[1 2]);
Acell = reshape(Acell,2,2);
desired = cell2mat(Acell)
desired =
1 1 3 3
1 1 3 3
2 2 4 4
2 2 4 4
C = permute(A,[1 3 2]);
C = reshape(C,[],4,1)
C =
1 3 1 3
1 3 1 3
2 4 2 4
2 4 2 4

Omar Mian
Omar Mian il 21 Mar 2011
Thank you Andrew & cyclist,
Both solutions work. Cyclist's permute solution is more efficient, taking approx approx 17% of the time used by the cell array solution.

Categorie

Scopri di più su Matrices and Arrays in Help Center e File Exchange

Prodotti

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by