Struct QuantileRegressor 

pub struct QuantileRegressor<F> {
    pub quantile: F,
    pub alpha: F,
    pub max_iter: usize,
    pub tol: F,
    pub fit_intercept: bool,
}

Quantile Regressor — conditional quantile estimation via IRLS.

Minimises the pinball loss with optional L1 regularization. The IRLS weights are w_i = 1 / (2 * max(|r_i|, eps)), which gives the iteratively reweighted least absolute deviations procedure.

Type Parameters

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

Fields

quantile: F

The quantile to estimate (must be in (0, 1)). Default 0.5 (median).

alpha: F

L1 regularization strength.

max_iter: usize

Maximum number of IRLS iterations.

tol: F

Convergence tolerance on the maximum coefficient change.

fit_intercept: bool

Whether to fit an intercept (bias) term.

Implementations

impl<F: Float + FromPrimitive> QuantileRegressor<F>

pub fn new() -> Self

Create a new QuantileRegressor with default settings.

Defaults: quantile = 0.5, alpha = 1.0, max_iter = 1000, tol = 1e-5, fit_intercept = true.

pub fn with_quantile(self, quantile: F) -> Self

Set the quantile to estimate.

Must be strictly between 0 and 1.

pub fn with_alpha(self, alpha: F) -> Self

Set the L1 regularization strength.

pub fn with_max_iter(self, max_iter: usize) -> Self

Set the maximum number of IRLS iterations.

pub fn with_tol(self, tol: F) -> Self

Set the convergence tolerance.

pub fn with_fit_intercept(self, fit_intercept: bool) -> Self

Set whether to fit an intercept term.

Trait Implementations

impl<F: Clone> Clone for QuantileRegressor<F>

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

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for QuantileRegressor<F>

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

Formats the value using the given formatter.

impl<F: Float + FromPrimitive> Default for QuantileRegressor<F>

fn default() -> Self

Returns the “default value” for a type.

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

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

Fit the quantile regression model via IRLS.

Errors

• FerroError::ShapeMismatch — sample count mismatch.
• FerroError::InsufficientSamples — zero samples.
• FerroError::InvalidParameter — quantile outside (0, 1) or negative alpha.

type Fitted = FittedQuantileRegressor<F>

The fitted model type returned by fit.

type Error = FerroError

The error type returned by fit.

impl<F> PipelineEstimator<F> for QuantileRegressor<F>

where F: Float + FromPrimitive + ScalarOperand + Send + Sync + 'static,

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

Fit this estimator on the given data.