Struct BayesianGaussianMixture 

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

Bayesian Gaussian Mixture Model

A Bayesian variant of Gaussian mixture model that uses variational inference to automatically determine the effective number of components. This implementation provides uncertainty quantification and automatic model selection capabilities.

The model uses variational Bayesian inference with proper priors on the mixture weights, means, and covariances to enable automatic component selection.

Parameters

• n_components - Maximum number of mixture components
• covariance_type - Type of covariance parameters (currently supports “full”)
• tol - Convergence threshold for the variational lower bound
• reg_covar - Regularization added to the diagonal of covariance
• max_iter - Maximum number of variational EM iterations
• random_state - Random state for reproducibility
• warm_start - Whether to use previous fit as initialization
• weight_concentration_prior_type - Type of prior on mixture weights
• weight_concentration_prior - Prior concentration parameter for mixture weights
• mean_precision_prior - Prior precision for component means
• mean_prior - Prior mean for component means
• degrees_of_freedom_prior - Prior degrees of freedom for covariance matrices
• covariance_prior - Prior scale for covariance matrices

Examples

use sklears_mixture::{BayesianGaussianMixture, 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 bgmm = BayesianGaussianMixture::new()
    .n_components(4)  // Will automatically select effective number
    .max_iter(100);
let fitted = bgmm.fit(&X.view(), &()).unwrap();
let labels = fitted.predict(&X.view()).unwrap();
println!("Effective components: {}", fitted.n_components_effective());

Implementations

impl BayesianGaussianMixture<Untrained>

pub fn new() -> Self

Create a new BayesianGaussianMixture instance

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

Set the maximum number of components

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

Set the covariance type

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

Set the convergence tolerance

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

Set the regularization parameter

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 warm_start(self, warm_start: bool) -> Self

Set warm start

pub fn weight_concentration_prior_type(self, prior_type: String) -> Self

Set weight concentration prior type

pub fn weight_concentration_prior(self, prior: f64) -> Self

Set weight concentration prior

pub fn mean_precision_prior(self, prior: f64) -> Self

Set mean precision prior

impl BayesianGaussianMixture<BayesianGaussianMixtureTrained>

pub fn weights(&self) -> &Array1<f64>

Get the mixture weights

pub fn means(&self) -> &Array2<f64>

Get the component means

pub fn covariances(&self) -> &[Array2<f64>]

Get the component covariances

pub fn n_components_effective(&self) -> usize

Get the effective number of components

pub fn lower_bound(&self) -> f64

Get the lower bound on the log likelihood

pub fn converged(&self) -> bool

Check if the algorithm converged

pub fn n_iter(&self) -> usize

Get the number of iterations performed

pub fn predict_proba(&self, X: &ArrayView2<'_, Float>) -> SklResult<Array2<f64>>

Predict probabilities for each component

pub fn score_samples(&self, X: &ArrayView2<'_, Float>) -> SklResult<Array1<f64>>

Compute the per-sample log-likelihood

pub fn score(&self, X: &ArrayView2<'_, Float>) -> SklResult<f64>

Compute the average log-likelihood

Trait Implementations

impl<S: Clone> Clone for BayesianGaussianMixture<S>

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

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<S: Debug> Debug for BayesianGaussianMixture<S>

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

Formats the value using the given formatter.

impl Default for BayesianGaussianMixture<Untrained>

fn default() -> Self

Returns the “default value” for a type.

impl Estimator for BayesianGaussianMixture<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 BayesianGaussianMixture<Untrained>

type Fitted = BayesianGaussianMixture<BayesianGaussianMixtureTrained>

The fitted model type

fn fit(self, X: &ArrayView2<'_, Float>, _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

impl Predict<ArrayBase<ViewRepr<&f64>, Dim<[usize; 2]>>, ArrayBase<OwnedRepr<i32>, Dim<[usize; 1]>>> for BayesianGaussianMixture<BayesianGaussianMixtureTrained>

fn predict(&self, X: &ArrayView2<'_, Float>) -> SklResult<Array1<i32>>

Make predictions on the provided data

fn predict_with_uncertainty( &self, x: &X, ) -> Result<(Output, UncertaintyMeasure), SklearsError>

Make predictions with confidence intervals