Struct NUTSSampler 

pub struct NUTSSampler { /* private fields */ }

No-U-Turn Sampler for Bayesian Mixture Models

NUTS is an adaptive variant of Hamiltonian Monte Carlo that automatically determines the number of leapfrog steps by continuing until the trajectory starts to “turn around” and move back towards its starting point. This makes it highly efficient for exploring complex posterior distributions without requiring manual tuning of the trajectory length.

Parameters

• n_components - Number of mixture components
• n_samples - Number of MCMC samples to draw
• n_warmup - Number of warmup samples for adaptation
• step_size - Initial step size for leapfrog integration
• target_accept_rate - Target acceptance rate for step size adaptation
• max_tree_depth - Maximum tree depth to prevent infinite loops
• adapt_step_size - Whether to adapt step size during warmup
• adapt_mass_matrix - Whether to adapt the mass matrix during warmup
• random_state - Random state for reproducibility

Examples

use sklears_mixture::{NUTSSampler, CovarianceType};
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 nuts = NUTSSampler::new()
    .n_components(2)
    .n_samples(1000)
    .n_warmup(500)
    .target_accept_rate(0.8)
    .max_tree_depth(10);

let result = nuts.sample(&X.view()).unwrap();

Implementations

impl NUTSSampler

pub fn new() -> Self

Create a new NUTSSampler instance

pub fn builder() -> Self

Create a new NUTSSampler instance using builder pattern (alias for new)

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

Set the number of components

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

Set the number of samples to draw

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

Set the number of warmup samples

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

Set the initial step size

pub fn target_accept_rate(self, rate: f64) -> Self

Set the target acceptance rate

pub fn max_tree_depth(self, depth: usize) -> Self

Set the maximum tree depth

pub fn adapt_step_size(self, adapt: bool) -> Self

Set whether to adapt step size

pub fn adapt_mass_matrix(self, adapt: bool) -> Self

Set whether to adapt mass matrix

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

Set the random state

pub fn build(self) -> Self

Build the NUTSSampler (builder pattern completion)

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

Sample from the posterior distribution of a Gaussian mixture model

Trait Implementations

impl Clone for NUTSSampler

fn clone(&self) -> NUTSSampler

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for NUTSSampler

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

Formats the value using the given formatter.

impl Default for NUTSSampler

fn default() -> Self

Returns the “default value” for a type.