DGLHeteroGraph.all_edges(form='uv', order='eid', etype=None)[source]

Return all edges with the specified edge type.

  • form (str, optional) –

    The return form, which can be one of the following:

    • 'eid': The returned result is a 1D tensor \(EID\), representing the IDs of all edges.

    • 'uv' (default): The returned result is a 2-tuple of 1D tensors \((U, V)\), representing the source and destination nodes of all edges. For each \(i\), \((U[i], V[i])\) forms an edge.

    • 'all': The returned result is a 3-tuple of 1D tensors \((U, V, EID)\), representing the source nodes, destination nodes and IDs of all edges. For each \(i\), \((U[i], V[i])\) forms an edge with ID \(EID[i]\).

  • order (str, optional) –

    The order of the returned edges, which can be one of the following:

    • 'srcdst': The edges are sorted first by their source node IDs and then by their destination node IDs to break ties.

    • 'eid' (default): The edges are sorted by their IDs.

  • etype (str or tuple of str, optional) – The edge type for query, which can be an edge type (str) or a canonical edge type (3-tuple of str). When an edge type appears in multiple canonical edge types, one must use a canonical edge type. If the graph has multiple edge types, one must specify the argument. Otherwise, it can be omitted.


All edges of the specified edge type. For a description of the returned result, see the description of form.

Return type

Tensor or (Tensor, Tensor) or (Tensor, Tensor, Tensor)


The following example uses PyTorch backend.

>>> import dgl
>>> import torch

Create a homogeneous graph.

>>> g = dgl.graph((torch.tensor([0, 0, 1, 1]), torch.tensor([1, 0, 2, 3])))

Query for edges.

>>> g.all_edges()
(tensor([0, 0, 1, 1]), tensor([1, 0, 2, 3]))

Specify a different value for form and order.

>>> g.all_edges(form='all', order='srcdst')
(tensor([0, 0, 1, 1]), tensor([0, 1, 2, 3]), tensor([1, 0, 2, 3]))

For a graph of multiple edge types, it is required to specify the edge type in query.

>>> hg = dgl.heterograph({
...     ('user', 'follows', 'user'): (torch.tensor([0, 1]), torch.tensor([1, 2])),
...     ('user', 'plays', 'game'): (torch.tensor([3, 4]), torch.tensor([5, 6]))
... })
>>> hg.all_edges(etype='plays')
(tensor([3, 4]), tensor([5, 6]))