geometric_edges#
- geometric_edges(G, radius, p=2)[source]#
Returns edge list of node pairs within
radius
of each other.- Parameters:
- Gnetworkx graph
The graph from which to generate the edge list. The nodes in
G
should have an attributepos
corresponding to the node position, which is used to compute the distance to other nodes.- radiusscalar
The distance threshold. Edges are included in the edge list if the distance between the two nodes is less than
radius
.- pscalar, default=2
The Minkowski distance metric used to compute distances. The default value is 2, i.e. Euclidean distance.
- Returns:
- edgeslist
List of edges whose distances are less than
radius
Notes
Radius uses Minkowski distance metric
p
. If scipy is available,scipy.spatial.cKDTree
is used to speed computation.Examples
Create a graph with nodes that have a “pos” attribute representing 2D coordinates.
>>> G = nx.Graph() >>> G.add_nodes_from([ ... (0, {"pos": (0, 0)}), ... (1, {"pos": (3, 0)}), ... (2, {"pos": (8, 0)}), ... ]) >>> nx.geometric_edges(G, radius=1) [] >>> nx.geometric_edges(G, radius=4) [(0, 1)] >>> nx.geometric_edges(G, radius=6) [(0, 1), (1, 2)] >>> nx.geometric_edges(G, radius=9) [(0, 1), (0, 2), (1, 2)]