fast_label_propagation_communities#

fast_label_propagation_communities(G, *, weight=None, seed=None)[source]#

Returns communities in G as detected by fast label propagation.

The fast label propagation algorithm is described in [1]. The algorithm is probabilistic and the found communities may vary in different executions.

The algorithm operates as follows. First, the community label of each node is set to a unique label. The algorithm then repeatedly updates the labels of the nodes to the most frequent label in their neighborhood. In case of ties, a random label is chosen from the most frequent labels.

The algorithm maintains a queue of nodes that still need to be processed. Initially, all nodes are added to the queue in a random order. Then the nodes are removed from the queue one by one and processed. If a node updates its label, all its neighbors that have a different label are added to the queue (if not already in the queue). The algorithm stops when the queue is empty.

Parameters:
GGraph, DiGraph, MultiGraph, or MultiDiGraph

Any NetworkX graph.

weightstring, or None (default)

The edge attribute representing a non-negative weight of an edge. If None, each edge is assumed to have weight one. The weight of an edge is used in determining the frequency with which a label appears among the neighbors of a node (edge with weight w is equivalent to w unweighted edges).

seedinteger, random_state, or None (default)

Indicator of random number generation state. See Randomness.

Returns:
communitiesiterable

Iterable of communities given as sets of nodes.

Notes

Edge directions are ignored for directed graphs. Edge weights must be non-negative numbers.

References

[1]

Vincent A. Traag & Lovro Šubelj. “Large network community detection by fast label propagation.” Scientific Reports 13 (2023): 2701. https://doi.org/10.1038/s41598-023-29610-z

Additional backends implement this function

neptuneNeptune computation backend for NetworkX.

  • The seed parameter is not supported at the moment.

  • Label propagation in Neptune Analytics maps all NetworkX variants to the same algorithm, using a fixed label update strategy. Variant-specific control over the update method (e.g., synchronous vs. asynchronous) is not configurable.

  • For additional parameters, see awslabs/nx-neptune