Struct MaximumLikelihoodEstimator 

pub struct MaximumLikelihoodEstimator {
    pub use_laplace: bool,
    pub pseudocount: f64,
}

Maximum Likelihood Estimator for discrete distributions.

Estimates parameters by counting frequencies in complete data.

Fields

use_laplace: bool

Use Laplace smoothing (add-one smoothing)

pseudocount: f64

Pseudocount for Laplace smoothing

Implementations

impl MaximumLikelihoodEstimator

pub fn new() -> Self

Create a new MLE estimator.

pub fn with_laplace(pseudocount: f64) -> Self

Create an MLE estimator with Laplace smoothing.

pub fn estimate_marginal( &self, variable: &str, cardinality: usize, data: &[Assignment], ) -> Result<ArrayD<f64>>

Estimate parameters for a single variable from data.

Arguments

• variable - Variable name
• cardinality - Number of possible values
• data - Observed assignments

Returns

Estimated probability distribution P(variable)

pub fn estimate_conditional( &self, child: &str, parents: &[String], cardinalities: &[usize], data: &[Assignment], ) -> Result<ArrayD<f64>>

Estimate conditional probability table P(child | parents) from data.

Arguments

• child - Child variable name
• parents - Parent variable names
• cardinalities - Cardinalities for [child, parent1, parent2, …]
• data - Observed assignments

Trait Implementations

impl Clone for MaximumLikelihoodEstimator

fn clone(&self) -> MaximumLikelihoodEstimator

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MaximumLikelihoodEstimator

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

Formats the value using the given formatter.

impl Default for MaximumLikelihoodEstimator

fn default() -> Self

Returns the “default value” for a type.