How to copy every 500 rows to a new variable?

1 visualización (últimos 30 días)
akshay raj
akshay raj el 16 de Nov. de 2015
Respondida: Walter Roberson el 19 de Jun. de 2019
Hi,
I have a [30000 x 57] matrix, I want to copy every new 500 rows to a new variable?
Thanks.

Respuesta aceptada

Stephen23
Stephen23 el 16 de Nov. de 2015
Editada: Stephen23 el 19 de Jun. de 2019
  1 comentario
Ilham Hardy
Ilham Hardy el 16 de Nov. de 2015
Wow, thanks for the concise explanation.
Now I am really convinced not to dynamically named variables.
Thanks Stephen :)

Iniciar sesión para comentar.

Más respuestas (4)

Star Strider
Star Strider el 16 de Nov. de 2015
Editada: Star Strider el 16 de Nov. de 2015
Another approach:
M = randi(9, 30000, 57); % Create Data
Mr = reshape(M, 50, [], 57); % Reshape
Mr = permute(Mr, [1 3 2]); % Shuffle Dimensions
Mc = mat2cell(Mr, ones(size(Mr,1),1), size(Mr,2), size(Mr,3)); % Create Cell Array
Every element of ‘Mc’, from ‘Mc{1}’ to ‘Mc{50}’ is a different matrix variable. To convert them back into double arrays for calculations, use the cell2mat function.
  2 comentarios
Bish Erbas
Bish Erbas el 16 de Nov. de 2015
Awesome!
Star Strider
Star Strider el 16 de Nov. de 2015
Thank you!
My code does the same essential task yours does, just differently.
That brings to focus the observation that neither you nor I created 50 ‘new’ variables with our code, but instead created a single cell array with segments of the original matrix that are easier to access and work with, as ‘akshay raj’ essentially requested.
It would seem that reading and understanding the existing Answers would avoid much of the consternation that is too frequently apparent otherwise.

Iniciar sesión para comentar.


Thorsten
Thorsten el 16 de Nov. de 2015
Editada: Thorsten el 16 de Nov. de 2015
Why do you want to create 60 new variables and double the data? That's not a good idea. You can keep your original data and just address the i'th set of your 500 rows using
ind = [1:500] + (i-1)*500;
your_function(X(ind,:));

Bish Erbas
Bish Erbas el 16 de Nov. de 2015
Do you really have to create individual variables for each 500 rows? This approach is not recommended because it is not using the array processing power of MATLAB. The preferred method is to store related data in a single array or a cell array. For example:
A = rand(30000,57);
N = (size(A,1)/500);
cellVar = cell(1,N);
for k = 1:N
cellVar{k} = A(500*(k-1)+1:500*(k-1)+500,:);
end

Walter Roberson
Walter Roberson el 19 de Jun. de 2019
result = squeeze(cellfun(@transpose, num2cell(reshape(YourMatrix.', size(YourMatrix,2), 500, []), [1 2]), 'uniform', 0));
This will give you a 60 x 1 cell array, each entry of which is a 500 x 57 array.

Categorías

Más información sobre Multidimensional Arrays en Help Center y File Exchange.

Etiquetas

Productos

Community Treasure Hunt

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

Start Hunting!

Translated by