Function asymmetric_preference_game 

pub fn asymmetric_preference_game(
    nodes: u32,
    no_out_types: u32,
    no_in_types: u32,
    type_dist_matrix: Option<&[Vec<f64>]>,
    pref_matrix: &[Vec<f64>],
    loops: bool,
    seed: u64,
) -> IgraphResult<(Graph, Vec<u32>, Vec<u32>)>

Generate a random directed graph with asymmetric vertex types and connection preferences.

Each vertex is assigned an (out_type, in_type) pair drawn jointly from type_dist_matrix (or uniformly when None). A directed edge u → v is then sampled with probability pref_matrix[out_type(u)][in_type(v)].

• nodes — number of vertices.
• no_out_types, no_in_types — number of out- and in-types (≥ 1).
• type_dist_matrix — out_types × in_types joint weights; None ⇒ uniform.
• pref_matrix — out_types × in_types connection probabilities in [0, 1].
• loops — allow self-loop edges.
• seed — initialises an internal SplitMix64 PRNG.

The resulting graph is always directed.

Returns

(graph, node_type_out, node_type_in) — the per-vertex out- and in-type assignments.

Errors

Returns IgraphError::InvalidArgument when:

• no_out_types < 1 or no_in_types < 1,
• type_dist_matrix has wrong dimensions, NaN, or negative entries,
• pref_matrix has wrong dimensions, NaN, or entries outside [0, 1],
• nodes exceeds the 2^53 f64-exact ceiling.

Examples

use rust_igraph::asymmetric_preference_game;

// 2 out-types × 2 in-types, uniform mixing, p ≈ 0.1 across the board.
let pref = vec![vec![0.1, 0.1], vec![0.1, 0.1]];
let (g, out, inp) =
    asymmetric_preference_game(40, 2, 2, None, &pref, true, 42).unwrap();
assert_eq!(g.vcount(), 40);
assert!(g.is_directed());
assert!(out.iter().all(|&t| t < 2));
assert!(inp.iter().all(|&t| t < 2));