Struct GradientBoostingRegressor 

pub struct GradientBoostingRegressor<F> {
    pub n_estimators: usize,
    pub learning_rate: f64,
    pub max_depth: Option<usize>,
    pub min_samples_split: usize,
    pub min_samples_leaf: usize,
    pub subsample: f64,
    pub loss: RegressionLoss,
    pub huber_alpha: f64,
    pub random_state: Option<u64>,
    /* private fields */
}

Gradient boosting regressor.

Builds an additive model in a forward stage-wise fashion, fitting each regression tree to the negative gradient of the loss function evaluated on the current ensemble prediction.

Type Parameters

• F: The floating-point type (f32 or f64).

Fields

n_estimators: usize

Number of boosting stages (trees).

learning_rate: f64

Learning rate (shrinkage) applied to each tree’s contribution.

max_depth: Option<usize>

Maximum depth of each tree.

min_samples_split: usize

Minimum number of samples required to split an internal node.

min_samples_leaf: usize

Minimum number of samples required in a leaf node.

subsample: f64

Fraction of samples to use for fitting each tree (stochastic boosting).

loss: RegressionLoss

Loss function.

huber_alpha: f64

Alpha quantile for Huber loss (only used when loss == Huber).

random_state: Option<u64>

Random seed for reproducibility.

Implementations

impl<F: Float> GradientBoostingRegressor<F>

pub fn new() -> Self

Create a new GradientBoostingRegressor with default settings.

Defaults: n_estimators = 100, learning_rate = 0.1, max_depth = Some(3), min_samples_split = 2, min_samples_leaf = 1, subsample = 1.0, loss = LeastSquares, huber_alpha = 0.9.

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

Set the number of boosting stages.

pub fn with_learning_rate(self, lr: f64) -> Self

Set the learning rate (shrinkage).

pub fn with_max_depth(self, d: Option<usize>) -> Self

Set the maximum tree depth.

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

Set the minimum number of samples to split a node.

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

Set the minimum number of samples in a leaf.

pub fn with_subsample(self, ratio: f64) -> Self

Set the subsample ratio (fraction of training data per tree).

pub fn with_loss(self, loss: RegressionLoss) -> Self

Set the loss function.

pub fn with_huber_alpha(self, alpha: f64) -> Self

Set the alpha quantile for Huber loss.

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

Set the random seed for reproducibility.

Trait Implementations

impl<F: Clone> Clone for GradientBoostingRegressor<F>

fn clone(&self) -> GradientBoostingRegressor<F>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for GradientBoostingRegressor<F>

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

Formats the value using the given formatter.

impl<F: Float> Default for GradientBoostingRegressor<F>

fn default() -> Self

Returns the “default value” for a type.

impl<F: Float + Send + Sync + 'static> Fit<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, ArrayBase<OwnedRepr<F>, Dim<[usize; 1]>>> for GradientBoostingRegressor<F>

fn fit( &self, x: &Array2<F>, y: &Array1<F>, ) -> Result<FittedGradientBoostingRegressor<F>, FerroError>

Fit the gradient boosting regressor.

Errors

Returns FerroError::ShapeMismatch if x and y have different numbers of samples. Returns FerroError::InsufficientSamples if there are no samples. Returns FerroError::InvalidParameter for invalid hyperparameters.

type Fitted = FittedGradientBoostingRegressor<F>

The fitted model type returned by fit.

type Error = FerroError

The error type returned by fit.

impl<F: Float + Send + Sync + 'static> PipelineEstimator<F> for GradientBoostingRegressor<F>

fn fit_pipeline( &self, x: &Array2<F>, y: &Array1<F>, ) -> Result<Box<dyn FittedPipelineEstimator<F>>, FerroError>

Fit this estimator on the given data.