Struct BayesianLinear 

pub struct BayesianLinear {
    pub mean: Vec<f64>,
    pub cov: Vec<f64>,
    pub beta: f64,
    pub fit_intercept: bool,
    pub p: usize,
}

A fitted Bayesian linear model: the posterior over weights (mean = m_N, cov = S_N) and the noise precision beta.

Fields

mean: Vec<f64>

cov: Vec<f64>

beta: f64

fit_intercept: bool

p: usize

Implementations

impl BayesianLinear

pub fn fit( x: &[f64], y: &[f64], n: usize, p: usize, alpha: f64, beta: f64, fit_intercept: bool, ) -> Result<Self, LearningError>

Fit with prior precision alpha (> 0) and noise precision beta (> 0). Fails closed on shape mismatch / non-positive hyper-parameters.

pub fn predict_row(&self, x_row: &[f64]) -> (f64, f64)

Predictive distribution at one row: (mean, variance). variance = 1/β + φᵀ S_N φ — the noise floor plus the model uncertainty, which grows away from the training data.

pub fn predict(&self, x: &[f64], m: usize) -> Vec<(f64, f64)>

Predictive (mean, variance) for each row of a row-major m × p matrix.

Trait Implementations

impl Clone for BayesianLinear

fn clone(&self) -> BayesianLinear

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for BayesianLinear

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

Formats the value using the given formatter.