How to create a multi-index vectors?
31 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
CaG
am 19 Feb. 2019
Bearbeitet: Stephen23
am 27 Mär. 2024 um 4:56
I have to create some vectors containing the values of a d-degree n-dimensional multi-index. A d-degree n-dimensional multi-index is a n-tuple such that .
Just to give an example, if I want a 2-degree 3-dimensional multi-index, I have to built the vectors:
I can to create them for the 2-dimensional case (basicly, I create a matrix and take the upper-right part), but when I move to higher dimensions I have no clue how to go on.
Do you have some suggestion?
0 Kommentare
Akzeptierte Antwort
Stephen23
am 19 Feb. 2019
Bearbeitet: Stephen23
am 19 Feb. 2019
Start by downloading John D'Errico's excellent partitions function:
and then using it like this:
d = 2;
n = 3;
P = partitions(d,1:d,n);
N = size(P,1);
C = cell(1,N);
for k = 1:N
tmp = repelem(1:d,P(k,:));
tmp(end+1:n) = 0;
C{k} = unique(perms(tmp),'rows');
end
Z = vertcat(C{:})
For d=2 and n=3 this gives:
Z =
0 1 1
1 0 1
1 1 0
0 0 2
0 2 0
2 0 0
For d=3 and n=4 this gives:
Z =
0 1 1 1
1 0 1 1
1 1 0 1
1 1 1 0
0 0 1 2
0 0 2 1
0 1 0 2
0 1 2 0
0 2 0 1
0 2 1 0
1 0 0 2
1 0 2 0
1 2 0 0
2 0 0 1
2 0 1 0
2 1 0 0
0 0 0 3
0 0 3 0
0 3 0 0
3 0 0 0
You might also be interested to read my answer here:
0 Kommentare
Weitere Antworten (1)
Firoozeh
am 26 Mär. 2024 um 22:03
Bearbeitet: Stephen23
am 27 Mär. 2024 um 4:29
function tensor = reconstruct_tensor(matricization_matrix, dimensions, n)
% Input:
% - matricization_matrix: The mode-n matricization matrix
% - dimensions: A vector containing the dimensions of the original tensor
% - n: The mode along which the matricization was performed
% Output:
% - tensor: The reconstructed tensor
% Validate input dimensions
if numel(dimensions) ~= length(dimensions)
error('Dimensions vector should be a 1D array.');
end
% Initialize the tensor
tensor = zeros(dimensions);
% Reshape the matricization matrix into the tensor
tensor = reshape(matricization_matrix, [dimensions(n), prod(dimensions) / dimensions(n)]);
% Permute dimensions to match the original order
perm_order = [n, setdiff(1:length(dimensions), n)];
tensor = permute(tensor, perm_order);
end
tensor = rand(3, 4, 5); % Example tensor
mode = 2; % Mode along which matricization was performed
matrix = matricize(tensor, mode); % Matricization of tensor
original_dimensions = [3, 4, 5]; % Original dimensions of tensor
reconstructed_tensor = reconstruct_tensor(matrix, original_dimensions, mode); % Reconstruct the tensor
Unrecognized function or variable 'reconstruct_tensor'.
what is the problem?
1 Kommentar
Stephen23
am 27 Mär. 2024 um 4:32
Bearbeitet: Stephen23
am 27 Mär. 2024 um 4:56
"what is the problem?"
There is no problem running the function here, so most likely you have not saved the function somewhere where MATLAB can see it (e.g. in the current directory, or as a local function at the end of a script).
mode = 2; % Mode along which matricization was performed
matrix = rand(3,4,5);
original_dimensions = [3, 4, 5]; % Original dimensions of tensor
reconstructed_tensor = reconstruct_tensor(matrix, original_dimensions, mode) % Reconstruct the tensor
function tensor = reconstruct_tensor(matricization_matrix, dimensions, n)
% Input:
% - matricization_matrix: The mode-n matricization matrix
% - dimensions: A vector containing the dimensions of the original tensor
% - n: The mode along which the matricization was performed
% Output:
% - tensor: The reconstructed tensor
% Validate input dimensions
if numel(dimensions) ~= length(dimensions)
error('Dimensions vector should be a 1D array.');
end
% Initialize the tensor
tensor = zeros(dimensions);
% Reshape the matricization matrix into the tensor
tensor = reshape(matricization_matrix, [dimensions(n), prod(dimensions) / dimensions(n)]);
% Permute dimensions to match the original order
perm_order = [n, setdiff(1:length(dimensions), n)];
tensor = permute(tensor, perm_order);
end
Siehe auch
Kategorien
Mehr zu Graphics Object Programming finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!