Reshape 2D matrix to 3D using time as 3rd dimension

1 visualización (últimos 30 días)
iijoclu
iijoclu el 28 de Oct. de 2015
Editada: Matt J el 12 de Mzo. de 2021
Hi,
I've got a A = n x n matrix and I want to convert this 2D matrix to a 3D one using time as the third dimension. So the 3D should show how the value of all the elements in the n x n matrix changes with respect to time.
I tried using the reshape function but to no avail.
Thanks!

Respuesta aceptada

Thorsten
Thorsten el 28 de Oct. de 2015
If i is your related to your time and takes values 1,2,3, etc, and A is the n x n matrix at time point i, you can use:
B(:,:,i) = A;
  2 comentarios
iijoclu
iijoclu el 28 de Oct. de 2015
Editada: iijoclu el 28 de Oct. de 2015
Ok so my time function is basically
[kgrid.t_array, dt] = makeTime(kgrid, medium.sound_speed);
When I try to plug kgrid.t_array in the formula above, I get an error saying that the subscripts should be real positive integers or logical.
Size of my kgrid.t_array is [1 5243].
Thorsten
Thorsten el 28 de Oct. de 2015
Editada: Thorsten el 28 de Oct. de 2015
You cannot use the values of kgrid.t_array, but instead 1:numel(kgrid.t_array). Your code should look something like below, assuming that you have a function that computes your NxN matrix for every timestep:
for i=1:numel(kgrid.t_array)
t = kgrid.t_array(i); % get the ith time value
A = computeNxNmatrix(t, more parameters); %
B(:,:,i) = A;
end

Iniciar sesión para comentar.

Más respuestas (1)

Matt J
Matt J el 28 de Oct. de 2015
I tried using the reshape function but to no avail.
Be more persistent. RESHAPE is the appropriate thing to use.
  2 comentarios
lmaree
lmaree el 12 de Mzo. de 2021
"Beyond the second dimension, the output, B, does not reflect trailing dimensions with a size of 1. For example, reshape(A,3,2,1,1) produces a 3-by-2 matrix." from the reshape documentation.
In my case, this is exactly what I need to multiply a spacial 2D matrix with a time vector. However, reshape doesn't have the functionality it seems...
Matt J
Matt J el 12 de Mzo. de 2021
Editada: Matt J el 12 de Mzo. de 2021
It is not needed. There is no difference between what you can do with a 3x2 matrix and with a 3x2x1x1 matrix.
A=rand(3,2);
t=rand(1,1,1,4);
A.*t
ans =
ans(:,:,1,1) = 0.2134 0.1069 0.0927 0.5706 0.2873 0.2506 ans(:,:,1,2) = 0.1807 0.0905 0.0785 0.4833 0.2433 0.2123 ans(:,:,1,3) = 0.3407 0.1707 0.1481 0.9111 0.4586 0.4002 ans(:,:,1,4) = 0.0462 0.0232 0.0201 0.1236 0.0622 0.0543

Iniciar sesión para comentar.

Categorías

Más información sobre Creating and Concatenating Matrices en Help Center y File Exchange.

Etiquetas

Community Treasure Hunt

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

Start Hunting!

Translated by