Struct GLM 

pub struct GLM { /* private fields */ }

Generalized Linear Model.

Fits regression models for exponential family distributions using IRLS (Iteratively Reweighted Least Squares).

Implementations

impl GLM

pub fn new(family: Family) -> Self

Creates a new GLM with the specified family and its canonical link.

pub fn with_link(self, link: Link) -> Self

Creates a GLM with a custom link function.

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

Sets the maximum number of IRLS iterations.

pub fn with_tolerance(self, tol: f32) -> Self

Sets the convergence tolerance.

pub fn with_dispersion(self, dispersion: f32) -> Self

Sets the dispersion parameter for Negative Binomial family.

The dispersion parameter α controls overdispersion: V(μ) = μ + α*μ². Higher α = more overdispersion. Default: 1.0.

pub fn coefficients(&self) -> Option<&[f32]>

Returns the fitted coefficients.

pub fn intercept(&self) -> Option<f32>

Returns the intercept.

pub fn fit(&mut self, x: &Matrix<f32>, y: &Vector<f32>) -> Result<()>

Fits the GLM using IRLS (Iteratively Reweighted Least Squares).

Arguments

• x - Feature matrix (n × p)
• y - Response vector (n)

Returns

Ok(()) on successful convergence, error otherwise

pub fn predict(&self, x_test: &Matrix<f32>) -> Result<Vector<f32>>

Predicts mean response for test data.

Returns μ = g⁻¹(Xβ + β₀).

Trait Implementations

impl Clone for GLM

fn clone(&self) -> GLM

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GLM

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

Formats the value using the given formatter.