Struct vrp_core::solver::population::Elitism

pub struct Elitism { /* fields omitted */ }

A simple evolution aware implementation of Population trait with the the following characteristics:

• sorting of individuals in population according their objective fitness using NSGA-II algorithm
• maintaining diversity of population based on their crowding distance

Implementations

impl Elitism

pub fn new(
    problem: Arc<Problem>,
    random: Arc<dyn Random + Send + Sync>,
    max_population_size: usize,
    selection_size: usize
) -> Self

Creates a new instance of Elitism.

• problem - a Vehicle Routing Problem definition.
• max_population_size - a max size of population size.

pub fn drain(&mut self) -> Vec<Individual>

Extracts all individuals out from population.

Trait Implementations

impl Display for Elitism

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

Formats the value using the given formatter.

impl Population for Elitism

fn add_all(&mut self, individuals: Vec<Individual>) -> bool

Adds all individuals into the population, then sorts and shrinks population if necessary. Returns true if any of newly added individuals is considered as best known.

fn add(&mut self, individual: Individual) -> bool

Adds an individual into the population. Returns true if newly added individual is considered as best known.

fn on_generation(&mut self, _: &Statistics)

Informs population about new generation event. This is time for the population to decide whether selection phase has to be changed.

fn cmp(&self, a: &Individual, b: &Individual) -> Ordering

Compares two solutions the same way as population does.

fn select<'a>(&'a self) -> Box<dyn Iterator<Item = &Individual> + 'a>

Selects parents from the population based on current selection phase.

fn ranked<'a>(&'a self) -> Box<dyn Iterator<Item = (&Individual, usize)> + 'a>

Returns subset of individuals within their rank sorted according their quality.

fn size(&self) -> usize

Returns population size.

fn selection_phase(&self) -> SelectionPhase

Returns a current selection phase.