Struct StructuredVariationalGMM 

pub struct StructuredVariationalGMM<S = Untrained> { /* private fields */ }

Structured Variational Gaussian Mixture Model

This implementation uses structured variational approximations that preserve certain dependencies between latent variables, providing more accurate posterior approximations than mean-field while remaining computationally tractable.

The key idea is to use a structured approximation of the form: q(θ, z) = q(π, μ, Λ | z) q(z) where some dependencies are preserved within each factor.

Examples

use sklears_mixture::{StructuredVariationalGMM, StructuredFamily, CovarianceType};
use sklears_core::traits::{Predict, Fit};
use scirs2_core::ndarray::array;

let X = array![[0.0, 0.0], [1.0, 1.0], [2.0, 2.0], [10.0, 10.0], [11.0, 11.0], [12.0, 12.0]];

let model = StructuredVariationalGMM::new()
    .n_components(2)
    .structured_family(StructuredFamily::MeanPrecision)
    .covariance_type(CovarianceType::Full);
let fitted = model.fit(&X.view(), &()).unwrap();
let labels = fitted.predict(&X.view()).unwrap();

Implementations

impl StructuredVariationalGMM<Untrained>

pub fn new() -> Self

Create a new structured variational GMM

pub fn n_components(self, n_components: usize) -> Self

Set the number of components

pub fn structured_family(self, family: StructuredFamily) -> Self

Set the structured approximation family

pub fn covariance_type(self, covariance_type: CovarianceType) -> Self

Set the covariance type

pub fn tol(self, tol: f64) -> Self

Set the convergence tolerance

pub fn max_iter(self, max_iter: usize) -> Self

Set the maximum number of iterations

pub fn random_state(self, random_state: u64) -> Self

Set the random state

pub fn reg_covar(self, reg_covar: f64) -> Self

Set the regularization parameter

pub fn weight_concentration(self, weight_concentration: f64) -> Self

Set the weight concentration parameter

pub fn mean_precision(self, mean_precision: f64) -> Self

Set the mean precision parameter

pub fn degrees_of_freedom(self, degrees_of_freedom: f64) -> Self

Set the degrees of freedom parameter

pub fn init_method(self, init_method: InitMethod) -> Self

Set the initialization method

pub fn n_init(self, n_init: usize) -> Self

Set the number of initializations

pub fn max_coord_steps(self, max_coord_steps: usize) -> Self

Set the maximum number of coordinate ascent steps

pub fn damping(self, damping: f64) -> Self

Set the damping factor

Trait Implementations

impl<S: Clone> Clone for StructuredVariationalGMM<S>

fn clone(&self) -> StructuredVariationalGMM<S>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<S: Debug> Debug for StructuredVariationalGMM<S>

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

Formats the value using the given formatter.

impl Default for StructuredVariationalGMM<Untrained>

fn default() -> Self

Returns the “default value” for a type.

impl Estimator for StructuredVariationalGMM<Untrained>

type Config = ()

Configuration type for the estimator

type Error = SklearsError

Error type for the estimator

type Float = f64

The numeric type used by this estimator

fn config(&self) -> &Self::Config

Get estimator configuration

fn validate_config(&self) -> Result<(), SklearsError>

Validate estimator configuration with detailed error context

fn check_compatibility( &self, n_samples: usize, n_features: usize, ) -> Result<(), SklearsError>

Check if estimator is compatible with given data dimensions

fn metadata(&self) -> EstimatorMetadata

Get estimator metadata

impl Fit<ArrayBase<ViewRepr<&f64>, Dim<[usize; 2]>>, ()> for StructuredVariationalGMM<Untrained>

type Fitted = StructuredVariationalGMMTrained

The fitted model type

fn fit(self, X: &ArrayView2<'_, f64>, _y: &()) -> SklResult<Self::Fitted>

Fit the model to the provided data with validation

fn fit_with_validation( self, x: &X, y: &Y, _x_val: Option<&X>, _y_val: Option<&Y>, ) -> Result<(Self::Fitted, FitMetrics), SklearsError>

where Self: Sized,

Fit with custom validation and early stopping