Struct RandomForest 

pub struct RandomForest { /* private fields */ }

Bagged ensemble of decision trees.

The forest stores full semantic Model trees rather than a bespoke forest node format. That costs some memory, but it keeps IR conversion, introspection, and optimized lowering consistent with the single-tree path.

Implementations

impl RandomForest

pub fn new( task: Task, criterion: Criterion, tree_type: TreeType, trees: Vec<Model>, compute_oob: bool, oob_score: Option<f64>, max_features: usize, seed: Option<u64>, num_features: usize, feature_preprocessing: Vec<FeaturePreprocessing>, ) -> Self

pub fn predict_table(&self, table: &dyn TableAccess) -> Vec<f64>

pub fn predict_proba_table( &self, table: &dyn TableAccess, ) -> Result<Vec<Vec<f64>>, PredictError>

pub fn task(&self) -> Task

pub fn criterion(&self) -> Criterion

pub fn tree_type(&self) -> TreeType

pub fn trees(&self) -> &[Model]

pub fn num_features(&self) -> usize

pub fn feature_preprocessing(&self) -> &[FeaturePreprocessing]

pub fn training_metadata(&self) -> TrainingMetadata

pub fn class_labels(&self) -> Option<Vec<f64>>

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

Trait Implementations

impl Clone for RandomForest

fn clone(&self) -> RandomForest

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RandomForest

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

Formats the value using the given formatter.