Struct MarkovModel 

pub struct MarkovModel { /* private fields */ }

Variable-order Markov model for sequence analysis

Implementations

impl MarkovModel

pub fn new(max_order: usize) -> AnomalyGridResult<Self>

Create a new Markov model with specified maximum order

pub fn with_config(config: AnomalyGridConfig) -> AnomalyGridResult<Self>

Create a new Markov model with custom configuration

pub fn train(&mut self, sequence: &[String]) -> AnomalyGridResult<()>

Train the model on a sequence

Complexity

• Time: O(n × max_order × |alphabet|) where n = sequence length
• Space: O(|alphabet|^max_order) in worst case

Performance Guarantees

• Memory usage is bounded by config.memory_limit if set
• Validates sequence length against config.min_sequence_length

pub fn calculate_likelihood(&self, sequence: &[String]) -> f64

Calculate the likelihood of a sequence under the model

pub fn get_best_context_probability( &self, context: &[String], next_state: &str, ) -> f64

Get the best context probability using adaptive hierarchical context selection

pub fn max_order(&self) -> usize

Get the maximum order of the model

pub fn config(&self) -> &AnomalyGridConfig

Get the configuration

pub fn state_mapping(&self) -> &HashMap<String, usize>

Get the state mapping

pub fn context_tree(&self) -> &ContextTree

Get the context tree

pub fn context_tree_mut(&mut self) -> &mut ContextTree

Get mutable access to the context tree (for optimizations)

pub fn get_marginal_probability(&self, state: &str) -> f64

Get marginal probability of a state from the training data

pub fn get_background_probability(&self, state: &str) -> f64

Get background probability for unseen transitions (public for adaptive scoring)

Trait Implementations

impl Clone for MarkovModel

fn clone(&self) -> MarkovModel

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MarkovModel

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

Formats the value using the given formatter.