Struct NormalEquationSolver 

pub struct NormalEquationSolver { /* private fields */ }

Normal Equations Solver (Streaming)

Solves $(A^T A) x = A^T b$.

• Pros: Low constant memory usage ($O(k^2)$ where $k$ is basis size). Parallelizable via summation.
• Cons: Numerically unstable if condition number is high ($\kappa(A^T A) = \kappa(A)^2$).

Implementations

impl NormalEquationSolver

pub fn new(n_basis: usize) -> Self

Trait Implementations

impl Clone for NormalEquationSolver

fn clone(&self) -> NormalEquationSolver

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for NormalEquationSolver

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

Formats the value using the given formatter.

impl RegressionSolver for NormalEquationSolver

fn create(n_basis: usize) -> Self

Creates a new solver instance initialized for n_basis features.

fn add_observation(&mut self, features: &[f64], target: f64)

Adds a single observation (row of A, element of b) to the solver.

fn solve(self, lambda: f64) -> Vec<f64>

Solves the regression problem and returns the coefficients. Consumes the solver state.

fn merge(&mut self, other: Self)

Merges another solver state into this one (for parallel reduction).