Struct Graph 

pub struct Graph {
    pub node_features: Vec<Vec<f64>>,
    pub adjacency: Vec<Vec<usize>>,
    pub labels: Option<Vec<usize>>,
}

A homogeneous graph with node feature vectors and adjacency lists.

Fields

node_features: Vec<Vec<f64>>

node_features[i] = feature vector for node i.

adjacency: Vec<Vec<usize>>

adjacency[i] = sorted list of neighbor indices for node i.

labels: Option<Vec<usize>>

Optional node-level class labels (for supervised training).

Implementations

impl Graph

pub fn new( node_features: Vec<Vec<f64>>, adjacency: Vec<Vec<usize>>, ) -> Result<Self>

Construct and validate a new graph.

pub fn with_labels(self, labels: Vec<usize>) -> Result<Self>

Attach labels (must match num_nodes()).

pub fn num_nodes(&self) -> usize

Number of nodes.

pub fn feature_dim(&self) -> usize

Feature dimension (0 if graph is empty).

pub fn neighbors(&self, v: usize) -> &[usize]

Get neighbors of node v.

pub fn sample_neighbors(&self, v: usize, k: usize, rng: &mut Lcg) -> Vec<usize>

Sample up to k neighbors uniformly without replacement.

Trait Implementations

impl Clone for Graph

fn clone(&self) -> Graph

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Graph

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.