Crate graphops 

graphops: graph operators (walk/diffusion + related centralities).

This crate is the canonical home for operator implementations in this workspace. Call sites should prefer graphops::* (and not refer to historical names).

Public invariants (must not drift):

• Node order: outputs are indexed by node id (0..n-1) consistent with the input graph’s adapter semantics (e.g. petgraph::NodeIndex::index() when using the petgraph feature).
• Determinism: deterministic operators are deterministic given identical inputs + configs.
• No silent normalization: normalization behavior is explicit in the API/docs (e.g. personalization vectors in PPR).

Swappable (allowed to change without breaking the contract):

• iteration strategy (serial vs parallel)
• convergence details (so long as tolerance semantics remain correct)
• internal data structures (so long as invariants hold)

Re-exports

pub use betweenness::betweenness_centrality;

pub use graph::PetgraphRef;

pub use graph::AdjacencyMatrix;

pub use graph::Graph;

pub use graph::GraphRef;

pub use graph::WeightedGraph;

pub use graph::WeightedGraphRef;

pub use node2vec::generate_biased_walks_precomp_ref;

pub use node2vec::generate_biased_walks_precomp_ref_from_nodes;

pub use node2vec::generate_biased_walks_weighted_plus_ref;

pub use node2vec::generate_biased_walks_weighted_ref;

pub use node2vec::PrecomputedBiasedWalks;

pub use node2vec::WeightedNode2VecPlusConfig;

pub use node2vec::generate_biased_walks_precomp_ref_parallel_from_nodes;

pub use random_walk::generate_biased_walks_ref_parallel;

pub use random_walk::generate_biased_walks_ref_parallel_from_nodes;

pub use random_walk::generate_walks_ref_parallel;

pub use random_walk::generate_walks_ref_parallel_from_nodes;

pub use centrality::closeness_centrality;

pub use centrality::harmonic_centrality;

pub use centrality::hits;

pub use eigenvector::eigenvector_centrality;

pub use eigenvector::eigenvector_centrality_run;

pub use eigenvector::EigenvectorRun;

pub use ellipsoidal::ellipsoid_distance;

pub use ellipsoidal::ellipsoid_overlap;

pub use ellipsoidal::ellipsoidal_embedding;

pub use ellipsoidal::Ellipsoid;

pub use ellipsoidal::EllipsoidalConfig;

pub use graph_kernel::random_walk_kernel;

pub use graph_kernel::sliced_wasserstein_graph_kernel;

pub use graph_kernel::structural_node_features;

pub use graph_kernel::wl_subtree_kernel;

pub use leiden::leiden;

pub use leiden::leiden_seeded;

pub use leiden::leiden_weighted;

pub use leiden::leiden_weighted_seeded;

pub use louvain::louvain;

pub use louvain::louvain_seeded;

pub use louvain::louvain_weighted;

pub use louvain::louvain_weighted_seeded;

pub use pagerank::pagerank;

pub use pagerank::pagerank_weighted;

pub use pagerank::PageRankConfig;

pub use pagerank::pagerank_checked;

pub use pagerank::pagerank_weighted_checked;

pub use pagerank::pagerank_checked_run;

pub use pagerank::pagerank_run;

pub use pagerank::pagerank_weighted_checked_run;

pub use pagerank::pagerank_weighted_run;

pub use pagerank::PageRankRun;

pub use partition::connected_components;

pub use partition::core_numbers;

pub use partition::k_core;

pub use partition::label_propagation;

pub use partition::strongly_connected_components;

pub use partition::topological_sort;

pub use ppr::personalized_pagerank;

pub use ppr::personalized_pagerank_checked;

pub use ppr::personalized_pagerank_checked_run;

pub use ppr::personalized_pagerank_run;

pub use random_walk::generate_biased_walks;

pub use random_walk::generate_biased_walks_from_nodes;

pub use random_walk::generate_biased_walks_ref;

pub use random_walk::generate_biased_walks_ref_from_nodes;

pub use random_walk::generate_biased_walks_ref_streaming_from_nodes;

pub use random_walk::generate_walks;

pub use random_walk::generate_walks_from_nodes;

pub use random_walk::generate_walks_ref;

pub use random_walk::generate_walks_ref_from_nodes;

pub use random_walk::generate_walks_ref_streaming_from_nodes;

pub use random_walk::sample_start_nodes_reservoir;

pub use random_walk::WalkConfig;

pub use reachability::reachability_counts_edges;

pub use shortest_path::bfs_distances;

pub use shortest_path::bfs_path;

pub use shortest_path::dijkstra_distances;

pub use similarity::cosine;

pub use similarity::jaccard;

pub use similarity::overlap;

pub use similarity::top_k_similar_jaccard;

pub use topk::normalize;

pub use topk::top_k;

pub use triangle::clustering_coefficients;

pub use triangle::global_clustering_coefficient;

pub use triangle::triangle_count;

Modules

betweenness

Betweenness centrality.

centrality

Centrality measures beyond PageRank/betweenness.

eigenvector

Eigenvector centrality via power iteration.

ellipsoidal

Ellipsoidal graph embeddings.

graph

Minimal graph adapter traits.

graph_kernel

Graph kernels: WL subtree, random walk, and sliced Wasserstein. Graph kernels: similarity measures between labeled graphs.

leiden

Leiden community detection (Traag, Waltman & van Eck, 2019).

louvain

Louvain community detection (Blondel et al., 2008).

node2vec

Node2Vec / Node2Vec+ walk generation (OTF + PreComp).

pagerank

PageRank centrality (power iteration on a random-surfer Markov chain).

partition

Graph partitioning / community detection.

ppr

Personalized PageRank.

random_walk

Random walk generation.

reachability

Reachability helpers (transitive closure counts).

shortest_path

Shortest path wrappers for GraphRef and WeightedGraphRef.

similarity

Node similarity based on neighborhood overlap.

topk

Ranking utilities.

triangle

Triangle counting and clustering coefficients.

Enums

Error

Errors returned by the *_checked variants of graph operators.

Type Aliases

Result

Shorthand for std::result::Result<T, Error>.