Function hsbm_list_game 

pub fn hsbm_list_game(
    n: u32,
    m_list: &[u32],
    rho_list: &[Vec<f64>],
    c_list: &[Vec<Vec<f64>>],
    p: f64,
    seed: u64,
) -> IgraphResult<Graph>

Generate a graph from the per-macro Hierarchical Stochastic Block Model: every macro-block carries its own size, its own micro-block proportions, and its own internal preference matrix.

• n — total vertex count. Must equal m_list.iter().sum().
• m_list — length-K vector of macro-block sizes. Each entry must be at least 1.
• rho_list — length-K list; entry b is the micro-block proportion vector for macro-block b. Each rho_list[b] must sum to 1 within √DBL_EPSILON, and every rho_list[b][j] * m_list[b] must be (within tolerance) an integer.
• c_list — length-K list; entry b is the k_b × k_b symmetric Bernoulli pref matrix for macro-block b. Entries in [0, 1].
• p — Bernoulli rate for every edge crossing two distinct macro-blocks. Must lie in [0, 1].
• seed — initialises an internal SplitMix64 PRNG.

The returned graph is always undirected, simple. Vertex ordering follows the macro-block order: macro b occupies [sum_{i<b} m_list[i], sum_{i<=b} m_list[i]).

Errors

Returns IgraphError::InvalidArgument when any of the constraints above is violated, including length mismatches between m_list, rho_list, and c_list.

Examples

use rust_igraph::hsbm_list_game;

// 3 macro-blocks of size 4, 6, 10. Each runs its own inner SBM.
let m_list = vec![4u32, 6, 10];
let rho_list = vec![
    vec![1.0],          // single cluster
    vec![0.5, 0.5],     // two equal clusters of 3
    vec![0.5, 0.5],     // two equal clusters of 5
];
let c_list = vec![
    vec![vec![0.5]],
    vec![vec![0.4, 0.1], vec![0.1, 0.4]],
    vec![vec![0.3, 0.05], vec![0.05, 0.3]],
];
let g = hsbm_list_game(20, &m_list, &rho_list, &c_list, 0.02, 7).unwrap();
assert_eq!(g.vcount(), 20);