dgl.sparse.SparseMatrix.coo

SparseMatrix.coo() → Tuple[torch.Tensor, torch.Tensor][source]

Returns the coordinate list (COO) representation of the sparse matrix.

See COO in Wikipedia.

Returns

  • torch.Tensor – Row coordinate

  • torch.Tensor – Column coordinate

Examples

>>> indices = torch.tensor([[1, 2, 1], [2, 4, 3]])
>>> A = from_coo(dst, src)
>>> A.coo()
(tensor([1, 2, 1]), tensor([2, 4, 3]))