Struct RandomForestRegressor 

pub struct RandomForestRegressor { /* private fields */ }

Random forest regressor (squared-error CART trees on bootstrap samples).

Defaults match scikit-learn’s RandomForestRegressor: 100 trees, all features per split, unlimited depth, min_samples_leaf = 1.

use dml_core::{MaxFeatures, RandomForestRegressor};
let rf = RandomForestRegressor::new()
    .n_estimators(200)
    .min_samples_leaf(5)
    .max_features(MaxFeatures::Fraction(0.5))
    .seed(42);

Implementations

impl RandomForestRegressor

pub fn new() -> Self

pub fn n_estimators(self, n: usize) -> Self

Number of trees (default 100).

pub fn max_depth(self, depth: Option<usize>) -> Self

Maximum tree depth (default: unlimited).

pub fn min_samples_split(self, n: usize) -> Self

Minimum number of rows required to split a node (default 2).

pub fn min_samples_leaf(self, n: usize) -> Self

Minimum number of distinct rows in each leaf (default 1). Values around 5–20 give smoother nuisance estimates for DML.

pub fn max_features(self, m: MaxFeatures) -> Self

Features considered per split.

pub fn bootstrap(self, bootstrap: bool) -> Self

Grow each tree on a bootstrap sample (default true).

pub fn max_samples(self, fraction: Option<f64>) -> Self

Bootstrap sample size as a fraction of n (default 1.0).

pub fn seed(self, seed: u64) -> Self

Seed for bootstrap draws and feature sampling (default 0).

pub fn feature_importances(&self) -> Option<&Array1<f64>>

Mean decrease in impurity per feature (sums to one), after fitting.

Trait Implementations

impl Clone for RandomForestRegressor

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 RandomForestRegressor

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

Formats the value using the given formatter.

impl Default for RandomForestRegressor

fn default() -> Self

Returns the “default value” for a type.

impl Regressor for RandomForestRegressor

fn fit(&mut self, x: ArrayView2<'_, f64>, y: ArrayView1<'_, f64>) -> Result<()>

fn fit_weighted( &mut self, x: ArrayView2<'_, f64>, y: ArrayView1<'_, f64>, w: ArrayView1<'_, f64>, ) -> Result<()>

Fit minimising a w-weighted loss (needed by the RLearner). The default reports that weights are unsupported.

fn predict(&self, x: ArrayView2<'_, f64>) -> Result<Array1<f64>>

fn fresh(&self) -> Box<dyn Regressor>