Struct BayesianHyperparameterLearner 

pub struct BayesianHyperparameterLearner {
    pub posteriors: HyperparameterPosteriors,
    /* private fields */
}

Online Bayesian hyperparameter learner

Fields

posteriors: HyperparameterPosteriors

Hyperparameter posteriors

Implementations

impl BayesianHyperparameterLearner

pub fn new() -> Self

Create a new learner

pub fn with_window_size(self, size: usize) -> Self

Set window size

pub fn with_decay(self, factor: f64) -> Self

Set decay factor (< 1.0 for non-stationary environments)

pub fn observe( &mut self, params: OperatorParams, parent_fitness: f64, child_fitness: f64, )

Observe the outcome of applying operators with given parameters

pub fn sample_params<R: Rng>(&self, rng: &mut R) -> OperatorParams

Sample parameters from current posteriors

pub fn map_params(&self) -> OperatorParams

Get MAP estimate of parameters

pub fn posteriors(&self) -> &HyperparameterPosteriors

Get current posteriors

pub fn average_improvement(&self) -> Option<f64>

Get average improvement in history

pub fn reset(&mut self)

Reset the learner

Trait Implementations

impl Clone for BayesianHyperparameterLearner

fn clone(&self) -> BayesianHyperparameterLearner

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for BayesianHyperparameterLearner

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

Formats the value using the given formatter.

impl Default for BayesianHyperparameterLearner

fn default() -> Self

Returns the “default value” for a type.