Struct NaturalGradientOptimizer 

pub struct NaturalGradientOptimizer<T: Float> { /* private fields */ }

Natural gradient optimizer for statistical manifolds

Implementations

impl<T: Float> NaturalGradientOptimizer<T>

pub fn new(config: NaturalGradientConfig<T>) -> Self

Create a new natural gradient optimizer with given configuration

pub fn with_default_config() -> Self

Create optimizer with default configuration

pub fn optimize_statistical<const DIM: usize, C: ConstraintState, O: ObjectiveState, V: ConvexityState>( &self, _problem: &OptimizationProblem<DIM, C, O, V, Statistical>, objective: &impl ObjectiveWithFisher<T>, initial_parameters: Vec<T>, ) -> OptimizationResult<NaturalGradientResult<T>>

Optimize on a statistical manifold

pub fn optimize_riemannian<const DIM: usize, C: ConstraintState, O: ObjectiveState, V: ConvexityState>( &self, _problem: &OptimizationProblem<DIM, C, O, V, Riemannian>, objective: &impl ObjectiveWithFisher<T>, initial_parameters: Vec<T>, ) -> OptimizationResult<NaturalGradientResult<T>>

Optimize on a Riemannian manifold

Trait Implementations

impl<T: Clone + Float> Clone for NaturalGradientOptimizer<T>

fn clone(&self) -> NaturalGradientOptimizer<T>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<T: Debug + Float> Debug for NaturalGradientOptimizer<T>

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

Formats the value using the given formatter.