Struct GaussianMixtureConfig 

pub struct GaussianMixtureConfig {
    pub max_iter: usize,
    pub loglik_tol: f64,
    pub parameter_tol: f64,
    pub covariance_floor: f64,
    pub kmeans_max_iter: usize,
}

Convergence + ladder controls for the discrete-mixture rung. All fields are fixed (no clock randomness, no env): deterministic seeding makes the fitted mixture a pure function of the data and k.

Fields

max_iter: usize

Exhaustion-escalation bound on EM iterations. It never selects the estimator: exhausting it without the convergence certificate (monotone ascent + relative objective step below loglik_tol) is an error, so an uncertified mixture never enters evidence comparison.

loglik_tol: f64

Relative mean-log-likelihood improvement tolerance for EM stopping.

parameter_tol: f64

Max-norm tolerance for the EM map in empirical predictive-density coordinates: the largest absolute change in any training row’s log density. This quotients component permutations, duplicate-component mass exchange, and non-identifiable ring factorizations while retaining sensitivity to likelihood changes that cancel in the mean objective.

covariance_floor: f64

Lower eigenvalue constraint for every component covariance. The M-step solves this constrained likelihood problem exactly by spectral clipping; it is not an additive ridge or an unmodelled prior.

kmeans_max_iter: usize

Maximum iterations for the deterministic k-means seeding pass.

Trait Implementations

impl Clone for GaussianMixtureConfig

fn clone(&self) -> GaussianMixtureConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for GaussianMixtureConfig

impl Debug for GaussianMixtureConfig

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

Formats the value using the given formatter.

impl Default for GaussianMixtureConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for GaussianMixtureConfig

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for GaussianMixtureConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.