Struct RandomSearch 

pub struct RandomSearch { /* private fields */ }

A search that draws every candidate uniformly.

Implementations

impl RandomSearch

pub fn new( space: SearchSpace, objective: Objective, max_evaluations: u64, seed: u64, ) -> Self

Returns a search of max_evaluations evaluations over space, drawing from seed.

Trait Implementations

impl Clone for RandomSearch

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RandomSearch

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

Formats the value using the given formatter.

impl Searcher for RandomSearch

fn ask(&mut self, max: usize) -> Vec<Candidate>

Returns up to max candidates to evaluate next.

fn tell(&mut self, evaluations: &[Evaluation])

Accepts the evaluations of the candidates that the last ask returned, sorted by candidate id.

fn is_done(&self) -> bool

Returns whether the search has issued its whole budget and been told every evaluation.

fn report(&self) -> SearchReport

Returns every candidate scored so far, best first, with the finished generations and the archive.

fn best(&self) -> Option<RankingEntry>

Returns the best candidate so far, or None before any evaluation or for a Pattern Space Exploration.

fn ranking_entry(&self, candidate_id: u64) -> Option<RankingEntry>

Returns candidate candidate_id with its objective over every replicate it has, or None for a candidate never told or a Pattern Space Exploration.

fn archive_entry(&self, _cell: PatternCell) -> Option<ArchiveEntry>

Returns the archive entry of cell, or None for an empty cell or a search other than a Pattern Space Exploration.

fn filled_cells(&self) -> u64

Number of filled cells in the archive of a Pattern Space Exploration, 0 for any other search.

fn generations(&self) -> &[GenerationSummary]

Finished generations of a genetic algorithm, in order, empty for any other search.

fn pattern_settings(&self) -> Option<&PatternSpaceSettings>

Returns the settings that a Pattern Space Exploration uses to place outputs, each axis with its range, or None while an automatic range waits for the initial samples or for any other search.