Enum HessianMethod 

#[non_exhaustive]
pub enum HessianMethod {
    GGN,
    Diagonal,
    KFAC,
}

Method used to approximate the Hessian of the loss for Laplace approximation.

• GGN: Generalized Gauss-Newton (Fisher information proxy) via squared gradients.
• Diagonal: Diagonal Hessian approximation.
• KFAC: Kronecker-factored approximate curvature.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

GGN

Generalized Gauss-Newton / squared-gradient Fisher approximation (default)

Diagonal

Diagonal curvature only

KFAC

Kronecker-factored approximate curvature

Trait Implementations

impl Clone for HessianMethod

fn clone(&self) -> HessianMethod

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for HessianMethod

impl Debug for HessianMethod

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

Formats the value using the given formatter.

impl Default for HessianMethod

fn default() -> HessianMethod

Returns the “default value” for a type.

impl Eq for HessianMethod

impl PartialEq for HessianMethod

fn eq(&self, other: &HessianMethod) -> 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 HessianMethod