dgl.sparse.SparseMatrix.cpu

SparseMatrix.cpu()[source]

Moves the matrix to CPU. If the matrix is already on CPU, the original matrix will be returned.

Returns:: The matrix on CPU
Return type:: SparseMatrix

Examples

>>> indices = torch.tensor([[1, 1, 2], [1, 2, 0]]).to("cuda")
>>> A = dglsp.spmatrix(indices, shape=(3, 4))

>>> A.cpu()
SparseMatrix(indices=tensor([[1, 1, 2],
                             [1, 2, 0]]),
             values=tensor([1., 1., 1.]),
             shape=(3, 4), nnz=3)