Enum RidgeSolver 

pub enum RidgeSolver {
    Auto,
    Cholesky,
    Svd,
}

Solver used for the dense single-output Ridge coefficient solve.

Mirrors a (dense) subset of scikit-learn’s solver parameter (sklearn/linear_model/_ridge.py:830 resolve_solver). Only the closed-form dense solvers are implemented here; the iterative solvers (lsqr/sparse_cg/sag/saga/lbfgs) are tracked separately as REQ-8b (blocker #386) and are intentionally NOT represented as variants until their iterative/SGD substrate lands.

Every dense solver returns the IDENTICAL ridge solution — the problem is strictly convex (alpha > 0, or alpha = 0 with full-rank X), so the minimizer is unique. The variant therefore only governs how the unique solution is computed, not what it is.

This governs only the single-output (Fit<Array2, Array1>) dense unconstrained solve. The multi-output (Fit<Array2, Array2>) path and the positive=true constrained path are unaffected by this setting.

Variants

Auto

'auto' — resolve automatically. For the dense path this resolves to Cholesky (sklearn resolve_solver, _ridge.py:830: dense, non-positive data → 'cholesky').

Cholesky

'cholesky' — solve the normal equations (XᵀX + alpha·I)·w = Xᵀy via a Cholesky factorization (_solve_cholesky, _ridge.py:741).

Svd

'svd' — solve via the singular value decomposition of X (_solve_svd, _ridge.py:200). Numerically identical to Cholesky on the strictly-convex ridge problem (unique minimizer).

Trait Implementations

impl Clone for RidgeSolver

fn clone(&self) -> RidgeSolver

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for RidgeSolver

impl Debug for RidgeSolver

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

Formats the value using the given formatter.

impl Default for RidgeSolver

fn default() -> RidgeSolver

Returns the “default value” for a type.

impl Eq for RidgeSolver

impl PartialEq for RidgeSolver

fn eq(&self, other: &RidgeSolver) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for RidgeSolver