dgl.sparse.diag

dgl.sparse.diag(val: Tensor, shape: Tuple[int, int] | None = None) → SparseMatrix[source]

Creates a sparse matrix based on the diagonal values.

Parameters:

val (torch.Tensor) – Diagonal of the matrix, in shape (N) or (N, D)
shape (tuple[int, int], optional) – If specified, len(val) must be equal to min(shape), otherwise, it will be inferred from val, i.e., (N, N)

Returns:

Sparse matrix

Return type:

SparseMatrix

Examples

Case1: 5-by-5 diagonal matrix with scaler values on the diagonal

>>> import torch
>>> val = torch.ones(5)
>>> dglsp.diag(val)
SparseMatrix(indices=tensor([[0, 1, 2, 3, 4],
                             [0, 1, 2, 3, 4]]),
             values=tensor([1., 1., 1., 1., 1.]),
             shape=(5, 5), nnz=5)

Case2: 5-by-10 diagonal matrix with scaler values on the diagonal

>>> val = torch.ones(5)
>>> dglsp.diag(val, shape=(5, 10))
SparseMatrix(indices=tensor([[0, 1, 2, 3, 4],
                             [0, 1, 2, 3, 4]]),
             values=tensor([1., 1., 1., 1., 1.]),
             shape=(5, 10), nnz=5)

Case3: 5-by-5 diagonal matrix with vector values on the diagonal

>>> val = torch.randn(5, 3)
>>> D = dglsp.diag(val)
>>> D.shape
(5, 5)
>>> D.nnz
5