Struct MaximumEntropyMarkovModel 

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

Maximum Entropy Markov Model (MEMM) for Sequence Labeling

MEMM is a discriminative model for sequence labeling that combines the advantages of maximum entropy models with Markov assumptions. Unlike CRF, MEMM models the conditional probability of each label given the previous label and observed features.

The model uses logistic regression at each position to predict the next label based on features extracted from the current observation and previous label.

Examples

use sklears_multioutput::MaximumEntropyMarkovModel;
use sklears_core::traits::{Predict, Fit};
// Use SciRS2-Core for arrays and random number generation (SciRS2 Policy)
use scirs2_core::ndarray::array;

// Sequence data: each row is a sequence element with features
let X = vec![
    array![[1.0, 2.0], [2.0, 3.0], [3.0, 1.0]], // sequence 1
    array![[4.0, 1.0], [1.0, 4.0]]                // sequence 2
];

// Label sequences
let y = vec![
    vec![0, 1, 0], // labels for sequence 1
    vec![1, 0]     // labels for sequence 2
];

let memm = MaximumEntropyMarkovModel::new()
    .max_iter(50)
    .learning_rate(0.01);
let trained_memm = memm.fit(&X, &y).unwrap();
let predictions = trained_memm.predict(&X).unwrap();

Implementations

impl MaximumEntropyMarkovModel<Untrained>

pub fn new() -> Self

Create a new MEMM instance

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

Set maximum number of iterations

pub fn learning_rate(self, learning_rate: Float) -> Self

Set learning rate

pub fn l2_regularization(self, l2_reg: Float) -> Self

Set L2 regularization strength

pub fn tolerance(self, tolerance: Float) -> Self

Set convergence tolerance

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

Set random state for reproducible results

impl MaximumEntropyMarkovModel<MaximumEntropyMarkovModelTrained>

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

Get the learned weights

pub fn n_labels(&self) -> usize

Get the number of labels

Trait Implementations

impl<S: Clone> Clone for MaximumEntropyMarkovModel<S>

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

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<S: Debug> Debug for MaximumEntropyMarkovModel<S>

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

Formats the value using the given formatter.

impl Default for MaximumEntropyMarkovModel<Untrained>

fn default() -> Self

Returns the “default value” for a type.

impl Estimator for MaximumEntropyMarkovModel<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<Vec<ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>>, Vec<Vec<i32>>> for MaximumEntropyMarkovModel<Untrained>

type Fitted = MaximumEntropyMarkovModel<MaximumEntropyMarkovModelTrained>

The fitted model type

fn fit( self, X: &Vec<Array2<Float>>, y: &Vec<Vec<i32>>, ) -> 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<Vec<ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>>, Vec<Vec<i32>>> for MaximumEntropyMarkovModel<MaximumEntropyMarkovModelTrained>

fn predict(&self, X: &Vec<Array2<Float>>) -> SklResult<Vec<Vec<i32>>>

Make predictions on the provided data

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

Make predictions with confidence intervals