Struct BaumWelchLearner 

pub struct BaumWelchLearner {
    pub max_iterations: usize,
    pub tolerance: f64,
    pub verbose: bool,
}

Baum-Welch algorithm for learning HMM parameters.

This is a specialized EM algorithm for Hidden Markov Models that learns:

• Initial state distribution
• Transition probabilities
• Emission probabilities

from sequences of observations (even when hidden states are not observed).

Fields

max_iterations: usize

Maximum number of EM iterations

tolerance: f64

Convergence tolerance (change in log-likelihood)

verbose: bool

Whether to print progress

Implementations

impl BaumWelchLearner

pub fn new(max_iterations: usize, tolerance: f64) -> Self

Create a new Baum-Welch learner.

pub fn with_verbose(max_iterations: usize, tolerance: f64) -> Self

Create a verbose learner that prints progress.

pub fn learn( &self, hmm: &mut SimpleHMM, observation_sequences: &[Vec<usize>], ) -> Result<f64>

Learn HMM parameters from observation sequences.

Arguments

• hmm - HMM model to update (will be modified in place)
• observation_sequences - Multiple observation sequences

Returns

Final log-likelihood

Trait Implementations

impl Clone for BaumWelchLearner

fn clone(&self) -> BaumWelchLearner

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for BaumWelchLearner

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

Formats the value using the given formatter.