Struct LinearRegression 

#[non_exhaustive]
pub struct LinearRegression { /* private fields */ }

Linear regression model.

Uses the OLS closed-form normal equations solution by default: β = (XᵀX + αI)⁻¹ Xᵀy. Set alpha > 0 for Ridge (L2) regularization.

Alternative solvers (QR, SVD) provide better numerical stability for ill-conditioned or rank-deficient problems.

When the gpu feature is enabled and the dataset is large enough, the XᵀX/Xᵀy computation is automatically offloaded to GPU compute shaders.

Implementations

impl LinearRegression

pub fn new() -> Self

Create a new linear regression model.

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

Set L2 regularization strength (Ridge regression).

pub fn solver(self, s: LinRegSolver) -> Self

Set the solver strategy.

pub fn fit(&mut self, data: &Dataset) -> Result<()>

Train the model on the given dataset.

pub fn predict(&self, features: &[Vec<f64>]) -> Result<Vec<f64>>

Predict target values.

pub fn fit_sparse(&mut self, features: &CscMatrix, target: &[f64]) -> Result<()>

Fit on sparse features (CSC format for column-oriented access).

Builds XᵀX and Xᵀy by iterating only over non-zero entries.

pub fn predict_sparse(&self, features: &CsrMatrix) -> Result<Vec<f64>>

Predict from sparse features (CSR format for row-oriented access).

pub fn coefficients(&self) -> &[f64]

Get learned coefficients.

pub fn intercept(&self) -> f64

Get learned intercept.

Trait Implementations

impl Clone for LinearRegression

fn clone(&self) -> LinearRegression

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Default for LinearRegression

fn default() -> Self

Returns the “default value” for a type.

impl PipelineModel for LinearRegression

fn fit(&mut self, data: &Dataset) -> Result<()>

Train the model on a dataset.

fn predict(&self, features: &[Vec<f64>]) -> Result<Vec<f64>>

Predict on row-major feature matrix.