Struct RandomForestSurrogate 

pub struct RandomForestSurrogate { /* private fields */ }

Random Forest surrogate for black-box optimization.

Maintains an ensemble of decision trees trained on bootstrap samples. The prediction mean and variance are computed as the mean and variance across the ensemble’s individual tree predictions.

Implementations

impl RandomForestSurrogate

pub fn new( n_estimators: usize, max_depth: usize, min_samples_split: usize, max_features: usize, seed: u64, ) -> Self

Create a new (unfitted) Random Forest surrogate.

pub fn default_config(seed: u64) -> Self

Default RF configuration (100 trees, depth 10, sqrt features).

pub fn fit(&mut self, x: &Array2<f64>, y: &Array1<f64>) -> OptimizeResult<()>

Fit the ensemble to training data.

x shape: (n_samples, n_features); y shape: (n_samples,).

pub fn predict( &self, x: &Array2<f64>, ) -> OptimizeResult<(Array1<f64>, Array1<f64>)>

Predict mean and variance from tree ensemble at test points.

Returns (mean, variance) where variance is the inter-tree variance.

pub fn predict_single(&self, x: &[f64]) -> OptimizeResult<(f64, f64)>

Predict at a single sample; returns (mean, std).

pub fn n_estimators(&self) -> usize

Number of trees in the ensemble.

pub fn is_fitted(&self) -> bool

Whether the model has been fitted.

Trait Implementations

impl Clone for RandomForestSurrogate

fn clone(&self) -> RandomForestSurrogate

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RandomForestSurrogate

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

Formats the value using the given formatter.