Struct NsgaIII 

pub struct NsgaIII {
    pub n_objectives: usize,
    pub population_size: usize,
    pub n_generations: usize,
    pub crossover_prob: f64,
    pub mutation_prob: f64,
    pub seed: u64,
    pub n_divisions: usize,
}

NSGA-III many-objective optimizer — struct-based API.

Example

use scirs2_optimize::multi_objective::advanced::NsgaIII;
let mut nsga = NsgaIII::new(2);
nsga.population_size = 20;
nsga.n_generations = 5;
let bounds: Vec<(f64, f64)> = vec![(0.0, 1.0); 3];
let result = nsga.optimize(
    |x| vec![x[0], 1.0 - x[0].sqrt()],
    &bounds,
).expect("valid input");
assert!(!result.pareto_front.is_empty());

Fields

n_objectives: usize

Number of objectives.

population_size: usize

Population size per generation.

n_generations: usize

Number of generations.

crossover_prob: f64

SBX crossover probability.

mutation_prob: f64

Polynomial mutation probability.

seed: u64

Random seed.

n_divisions: usize

Number of reference-point divisions per objective.

Implementations

impl NsgaIII

pub fn new(n_objectives: usize) -> Self

Create a new NSGA-III optimizer for the given number of objectives.

pub fn optimize<F>( &self, objectives: F, bounds: &[(f64, f64)], ) -> OptimizeResult<NsgaResult>

where F: Fn(&[f64]) -> Vec<f64>,

Run NSGA-III to approximate the Pareto front.

Trait Implementations

impl Clone for NsgaIII

fn clone(&self) -> NsgaIII

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for NsgaIII

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

Formats the value using the given formatter.