Enum RidgePolicy 

pub enum RidgePolicy {
    ExactFullObjective,
    SolverOnly,
}

Structurally valid ways a diagonal ridge may participate in a computation.

The former public boolean matrix admitted contradictory states such as a quadratic penalty without the corresponding Hessian. This enum has only the coherent inhabitants the engine actually selects.

Why there is no approximate-determinant inhabitant (#2670)

A third variant, PositivePartApproximateObjective, used to route the ridged log-determinant through a smooth positive-part spectral approximation (log|A|_reg = Σ log r_ε(σ_j)), which is a DIFFERENT estimand from the exact SPD determinant and was named as such. Nothing in production ever selected it: every construction of it lived under #[cfg(test)], so the only thing it bought the library was a second, worse answer a user could opt into by mistake. It is deleted rather than kept as a fallback — a preserved fallback is a second implementation of the same quantity, and this one changed the estimand while doing it.

The smooth regulariser itself is NOT deleted and was never this enum’s business: spectral_regularize / spectral_epsilon stay live in the REML outer engine’s DenseSpectralOperator, where a caller genuinely wants the smooth surrogate together with its matching analytic gradient.

With one determinant semantics left there is no determinant_mode() and no RidgeDeterminantMode: a query whose answer is a constant is not a query.

Variants

ExactFullObjective

Ridge is an explicit part of the exact objective: quadratic, penalty normalizer, and Laplace Hessian all include it, using a full SPD logdet.

SolverOnly

Ridge changes only an inner linear solve and never the fitted objective, exported Hessian, determinant, covariance, or serialized model.

Implementations

impl RidgePolicy

pub const fn exact_full_objective() -> RidgePolicy

pub const fn solver_only() -> RidgePolicy

pub const fn accounts_for_objective(self) -> bool

Trait Implementations

impl Clone for RidgePolicy

fn clone(&self) -> RidgePolicy

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for RidgePolicy

impl Debug for RidgePolicy

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for RidgePolicy

fn deserialize<__D>( __deserializer: __D, ) -> Result<RidgePolicy, <__D as Deserializer<'de>>::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for RidgePolicy

impl Hash for RidgePolicy

fn hash<__H>(&self, state: &mut __H)

where __H: Hasher,

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for RidgePolicy

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

Equality operator ==.

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

Inequality operator !=.

impl Serialize for RidgePolicy

fn serialize<__S>( &self, __serializer: __S, ) -> Result<<__S as Serializer>::Ok, <__S as Serializer>::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for RidgePolicy