Struct UnivariateGaussianState 

pub struct UnivariateGaussianState { /* private fields */ }

Fitted statistics for the UnivariateGaussian model after one call to ProbabilityModel::fit.

Both vectors have length genome_dim. On the prior path (prev = None) they are initialised from UnivariateGaussianParams::init_mean / init_std²; on subsequent calls they hold the unweighted MLE estimates computed from the truncation-selected population.

Fields are private so a mismatched mean / variance length (or a negative / non-finite variance) is unrepresentable from outside this module; use try_new to build one and the mean / variance accessors to read it.

Implementations

impl UnivariateGaussianState

pub fn try_new(mean: Vec<f32>, variance: Vec<f32>) -> Result<Self, ConfigError>

Builds a fitted Gaussian state from per-dimension mean and variance.

Errors

Returns a ConfigError if mean is empty, if mean and variance differ in length, or if any variance is negative or non-finite. A variance is a squared deviation, so it must be >= 0 and finite.

pub fn mean(&self) -> &[f32]

Per-dimension MLE means, length genome_dim.

pub fn variance(&self) -> &[f32]

Per-dimension MLE variances, length genome_dim.

Trait Implementations

impl Clone for UnivariateGaussianState

fn clone(&self) -> UnivariateGaussianState

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for UnivariateGaussianState

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

Formats the value using the given formatter.