cugraph

NVIDIA cuGraph is a graph analytics library that is part of NVIDIA’s RAPIDS open-source data science suite containing machine learning tools and libraries for various applications in data science; it can be used from Memgraph on machines that meet the system requirements.

This set of modules is built on top of NVIDIA cuGraph and provides a set of wrappers for most of the algorithms present in the cuGraph repository.

Trait	Value
Module type	module
Implementation	CUDA
Graph direction	undirected/directed
Edge weights	unweighted/weighted
Parallelism	parallelized

Modules

info

The cugraph module is a collection of distinct GPU-powered modules with their own procedures.

cugraph.balanced_cut_clustering

Procedures

info

If you want to execute this algorithm on graph projections, subgraphs or portions of the graph, be sure to check out the guide on How to run a MAGE module on subgraphs.

get(num_clusters, num_eigenvectors, ev_tolerance, ev_max_iter, kmean_tolerance, kmean_max_iter, weight_property)

Find the balanced cut clustering of the graph’s nodes.

Input:

• num_clusters: integer ➡ Number of clusters.
• num_eigenvectors: integer (default=2) ➡ Number of eigenvectors to be used (must be less than or equal to num_clusters).
• ev_tolerance: float (default=0.00001) ➡ Tolerance used by the eigensolver.
• ev_max_iter: integer (default=100) ➡ Maximum number of iterations for the eigensolver.
• kmean_tolerance: float (default=0.00001) ➡ Tolerance used by the k-means solver.
• kmean_max_iter: integer (default=100) ➡ Maximum number of iterations for the k-means solver.
• weight_property: string (default="weight") ➡ The values of the given relationship. property are used as weights by the algorithm. If this property is not set for a relationship, the fallback value is 1.0.

Output:

• node: Vertex ➡ Graph node.
• cluster: integer ➡ Cluster of a node.

Usage:

CALL cugraph.balanced_cut_clustering.get(3)
YIELD node, cluster
RETURN node, cluster;

cugraph.betweenness_centrality

Procedures

get(normalized, directed, weight_property)

Find betweenness centrality scores for all nodes in the graph.

Input:

• normalized: boolean (default=True) ➡ Normalize the output.
• directed: boolean (default=True) ➡ Graph directedness. (default True)
• weight_property: string (default="weight") ➡ The values of the given relationship property are used as weights by the algorithm. If this property is not set for a relationship, the fallback value is 1.0.

Output:

• node: Vertex ➡ Graph node.
• betweenness_centrality: float ➡ Betweenness centrality score of a node.

Usage:

CALL cugraph.betweenness_centrality.get()
YIELD node, betweenness_centrality
RETURN node, betweenness_centrality;

cugraph.generator

Procedures

rmat(scale, num_edges, node_labels, edge_type, a, b, c, seed, clip_and_flip)

Generate a graph using a Recursive MATrix (R-MAT) graph generation algorithm and load it in Memgraph.

Input:

• scale: integer (default=4) ➡ Scale factor to set the number of vertices in the graph.
• num_edges: integer (default=100) ➡ Number of edges in the generated graph.
• node_labels: mgp.List[string] ([]) ➡ Labels on created vertices. Defaults to empty list.
• edge_type: string (default="RELATIONSHIP") ➡ Edge type, defines the name of the relationship.
• a: double (default=0.57) ➡ First partition probability.
• b: double (default=0.19) ➡ Second partition probability.
• c: double (default=0.19) ➡ Third partition probability.
• seed: integer (default=0) ➡ RNG (random number generator) seed value
• clip_and_flip: boolean (default=False) ➡ Controls whether to generate edges only in the lower triangular part (including the diagonal) of the graph adjacency matrix (if set to True) or not (if set to False).

Output:

The generated graph is loaded into Memgraph.

• message: string ➡ Success message if the graph is loaded.

Usage:

CALL cugraph.generator.rmat() YIELD *;

cugraph.hits

Procedures

get(tolerance, max_iterations, normalized, directed)

Find HITS authority and hub values for all nodes in the graph. The HITS algorithm computes two numbers for each node: its authority, which estimates the value of its content, and its hub value, which estimates the value of its links to other nodes.

Whereas the HITS algorithm was designed for directed graphs, this implementation does not check if the input graph is directed and will execute on undirected graphs.

Input:

• tolerance: float (default=1e-5) ➡ HITS approximation tolerance (custom values not supported by NVIDIA cuGraph).
• max_iterations: integer (default=100) ➡ Maximum number of iterations before returning an answer (custom values not supported by NVIDIA cuGraph).
• normalized: boolean (default=True) ➡ Normalize the output (False not supported by NVIDIA cuGraph).
• directed: boolean (default=True) ➡ Graph directedness. (default True)

Output:

• node: Vertex ➡ Graph node.
• hubs: float ➡ Hub value of a node.
• authorities: float ➡ Authority value of a node.

Usage:

CALL cugraph.hits.get()
YIELD node, hubs, authorities
RETURN node, hubs, authorities;

cugraph.katz_centrality

Procedures

get(alpha, beta, epsilon, max_iterations, normalized, directed)

Find Katz centrality scores for all nodes in the graph.

Input:

• alpha: float (default=None) ➡ Attenuation factor defining the walk length importance. If not specified, calculated as 1 / max(out_degree).
• beta: float (default=1.0) ➡ Weight scalar (currently not supported by NVIDIA cuGraph).
• epsilon: float (default=1e-6) ➡ Set the tolerance for the approximation, this parameter should be a small magnitude value.
• max_iterations: integer (default=100) ➡ Maximum number of iterations before returning an answer.
• normalized: boolean (default=True) ➡ Normalize the output.
• directed: boolean (default=True) ➡ Graph directedness. (default True)

Output:

• node: Vertex ➡ Graph node.
• katz_centrality: float ➡ Katz centrality score of a node.

Usage:

CALL cugraph.katz_centrality.get()
YIELD node, katz_centrality
RETURN node, katz_centrality;

cugraph.leiden

Procedures

get(max_iterations, resolution)

Find the partition of the graph into communities using the Leiden method.

Input:

• max_iterations: integer (default=100) ➡ Maximum number of iterations (levels) of the algorithm.
• resolution: float (default=1.0) ➡ Controls community size (lower values lead to fewer, larger communities and vice versa).

Output:

• node: Vertex ➡ Graph node.
• partition: integer ➡ Partition of a node.

Usage:

CALL cugraph.leiden.get()
YIELD node, partition
RETURN node, partition;

cugraph.louvain

Procedures

get(max_iterations, resolution, directed)

Find the partition of the graph into communities using the Louvain method.

Input:

• max_iterations: integer (default=100) ➡ Maximum number of iterations (levels) of the algorithm.
• resolution: float (default=1.0) ➡ Controls community size (lower values lead to fewer, larger communities and vice versa).
• directed: boolean (default=True) ➡ Graph directedness. (default True)

Output:

• node: Vertex ➡ Graph node.
• partition: integer ➡ Partition of a node.

Usage:

CALL cugraph.louvain.get()
YIELD node, partition
RETURN node, partition;

cugraph.pagerank

Procedures

get(max_iterations, damping_factor, stop_epsilon, weight_property)

Find PageRank scores for all nodes in the graph.

Input:

• max_iterations: integer (default=100) ➡ The maximum number of iterations before returning an answer (default 100). Use it to limit the execution time or do an early exit before the solver reaches the convergence tolerance.
• damping_factor: float (default=0.85) ➡ The damping factor represents the probability to follow an outgoing edge.
• stop_epsilon: float (default=1e-5) ➡ The convergence tolerance for PageRank approximation. Lowering tolerance improves the approximation, but setting this parameter too low can ensue in non-convergence due to numerical round-off. Values between 0.01 and 0.00001 are usually acceptable.
• weight_property: string (default="weight") ➡ The values of the given relationship property are used as weights by the algorithm. If this property is not set for a relationship, the fallback value is 1.0.

Output:

• node: Vertex ➡ Graph node.
• pagerank: float ➡ PageRank score of a node.

Usage:

CALL cugraph.pagerank.get()
YIELD node, pagerank
RETURN node, pagerank;

cugraph.personalized_pagerank

Procedures

get(personalization_vertices, personalization_values, max_iterations, damping_factor, stop_epsilon, weight_property)

Find personalized PageRank scores for all nodes in the graph.

Input:

• personalization_vertices: mgp.List[mgp.Vertex] ➡ Graph nodes with personalization values.
• personalization_values: mgp.List[float] ➡ Above nodes’ personalization values.
• weight_property: string (default="weight") ➡ The values of the given relationship. property are used as weights by the algorithm. If this property is not set for a relationship, the fallback value is 1.0.
• damping_factor: float (default=0.85) ➡ The damping factor represents the probability to follow an outgoing edge.
• stop_epsilon: float (default=1e-5) ➡ The convergence tolerance for PageRank approximation. Lowering tolerance improves the approximation, but setting this parameter too low can ensue in non-convergence due to numerical round-off. Values between 0.01 and 0.00001 are usually acceptable.
• max_iterations: integer (default=100) ➡ The maximum number of iterations before returning an answer (default 100). Use it to limit the execution time or do an early exit before the solver reaches the convergence tolerance.

Output:

• node: Vertex ➡ Graph node.
• pagerank: float ➡ PageRank score of a node.

Usage:

MATCH (n: Node {id: 1}), (m: Node {id: 2})
CALL cugraph.pagerank.get([n, m], [0.2, 0.5])
YIELD node, pagerank
RETURN node, pagerank;

cugraph.spectral_clustering

Procedures

get(num_clusters, num_eigenvectors, ev_tolerance, ev_max_iter, kmean_tolerance, kmean_max_iter, weight_property)

Find the spectral clustering of the graph’s nodes.

Input:

• num_clusters: integer ➡ Number of clusters.
• num_eigenvectors: integer (default=2) ➡ Number of eigenvectors to be used (must be less than or equal to num_clusters).
• ev_tolerance: float (default=0.00001) ➡ Tolerance used by the eigensolver.
• ev_max_iter: integer (default=100) ➡ Maximum number of iterations for the eigensolver.
• kmean_tolerance: float (default=0.00001) ➡ Tolerance used by the k-means solver.
• kmean_max_iter: integer (default=100) ➡ Maximum number of iterations for the k-means solver.
• weight_property: string (default="weight") ➡ The values of the given relationship property are used as weights by the algorithm. If this property is not set for a relationship, the fallback value is 1.0.

Output:

• node: Vertex ➡ Graph node.
• cluster: integer ➡ Cluster of a node.

Usage:

CALL cugraph.spectral_clustering.get(3)
YIELD node, cluster
RETURN node, cluster;