Struct rusty_machine::learning::naive_bayes::NaiveBayes

pub struct NaiveBayes<T: Distribution> { /* private fields */ }

The Naive Bayes model.

Implementations

impl<T: Distribution> NaiveBayes<T>

pub fn new() -> NaiveBayes<T>

Create a new NaiveBayes model from a given distribution.

Examples

use rusty_machine::learning::naive_bayes::{NaiveBayes, Gaussian};

// Create a new Gaussian Naive Bayes model.
let _ = NaiveBayes::<Gaussian>::new();

pub fn cluster_count(&self) -> Option<&usize>

Get the cluster count for this model.

Returns an option which is None until the model has been trained.

pub fn class_prior(&self) -> Option<&Vec<f64>>

Get the class prior distribution for this model.

Returns an option which is None until the model has been trained.

pub fn distr(&self) -> Option<&T>

Get the distribution for this model.

Returns an option which is None until the model has been trained.

impl<T: Distribution> NaiveBayes<T>

pub fn get_log_probs(&self, inputs: &Matrix<f64>) -> LearningResult<Matrix<f64>>

Get the log-probabilities per class for each input.

Trait Implementations

impl<T: Debug + Distribution> Debug for NaiveBayes<T>

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

Formats the value using the given formatter.

impl<T: Default + Distribution> Default for NaiveBayes<T>

fn default() -> NaiveBayes<T>

Returns the “default value” for a type.

impl<T: Distribution> SupModel<Matrix<f64>, Matrix<f64>> for NaiveBayes<T>

Train and predict from the Naive Bayes model.

The input matrix must be rows made up of features. The target matrix should have indicator vectors in each row specifying the input class. e.g. [[1,0,0],[0,0,1]] shows class 1 first, then class 3.

fn train(
    &mut self,
    inputs: &Matrix<f64>,
    targets: &Matrix<f64>
) -> LearningResult<()>

Train the model using inputs and targets.

fn predict(&self, inputs: &Matrix<f64>) -> LearningResult<Matrix<f64>>

Predict output from inputs.