Struct StatisticalManifold 

pub struct StatisticalManifold {
    pub dim: usize,
}

A statistical manifold: a smooth manifold whose points are probability distributions parameterized by dim real parameters.

Provides the Fisher information metric, geodesics, and Christoffel symbols.

Fields

dim: usize

Dimension of the parameter space.

Implementations

impl StatisticalManifold

pub fn new(dim: usize) -> Self

Create a new StatisticalManifold of the given dimension.

pub fn fisher_metric(&self, params: &[f64]) -> Vec<Vec<f64>>

Compute the Fisher information metric (matrix) at params.

Uses a finite-difference approximation of the log-likelihood Hessian. Returns a dim × dim positive-semidefinite matrix.

pub fn geodesic(&self, p: &[f64], q: &[f64], t: f64) -> Vec<f64>

Compute the geodesic between parameter points p and q at time t ∈ [0,1].

Uses the exponential map with the Fisher metric (first-order approximation).

pub fn christoffel_symbols(&self, params: &[f64]) -> Vec<Vec<Vec<f64>>>

Compute the Christoffel symbols Γ^k_{ij} at params.

Uses a finite-difference approximation of the metric derivatives. Returns [k][i][j] indexed array.

Trait Implementations

impl Clone for StatisticalManifold

fn clone(&self) -> StatisticalManifold

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for StatisticalManifold

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

Formats the value using the given formatter.