Community

EgoNet

cugraph.batched_ego_graphs(G, seeds[, ...])	This function is deprecated.
cugraph.ego_graph(G, n[, radius, center, ...])	Compute the induced subgraph of neighbors centered at node n, within a given radius.

Ensemble clustering for graphs (ECG)

cugraph.ecg(input_graph[, min_weight, ...])

Compute the Ensemble Clustering for Graphs (ECG) partition of the input graph.

K-Truss

cugraph.k_truss(G, k)	Returns the K-Truss subgraph of a graph for a specific k.
cugraph.ktruss_subgraph(G, k[, use_weights])	Returns the K-Truss subgraph of a graph for a specific k.

Leiden

cugraph.leiden(G[, max_iter, resolution, ...])

Compute the modularity optimizing partition of the input graph using the Leiden algorithm

Louvain

cugraph.louvain(G[, max_level, max_iter, ...])

Compute the modularity optimizing partition of the input graph using the Louvain method

Louvain (MG)

cugraph.dask.community.louvain.louvain(...)

Compute the modularity optimizing partition of the input graph using the Louvain method

Spectral Clustering

cugraph.analyzeClustering_edge_cut(G, ...[, ...])	Compute the edge cut score for a partitioning/clustering The assumption is that “clustering” is the results from a call from a special clustering algorithm and contains columns named “vertex” and “cluster”.
cugraph.analyzeClustering_modularity(G, ...)	Compute the modularity score for a given partitioning/clustering.
cugraph.analyzeClustering_ratio_cut(G, ...)	Compute the ratio cut score for a partitioning/clustering
cugraph.spectralBalancedCutClustering(G, ...)	Compute a clustering/partitioning of the given graph using the spectral balanced cut method.
cugraph.spectralModularityMaximizationClustering(G, ...)	Compute a clustering/partitioning of the given graph using the spectral modularity maximization method.

Subgraph Extraction

cugraph.subgraph(G, vertices)

Compute a subgraph of the existing graph including only the specified vertices.

Triangle Counting

cugraph.triangle_count(G[, start_list])

Compute the number of triangles (cycles of length three) in the input graph.