Struct StdNaiveBayes 

pub struct StdNaiveBayes<State = Unfit> {
    pub alpha: f64,
    pub probability_of_class: HashMap<String, f64>,
    pub probability_of_feat_by_class: HashMap<String, HashMap<String, f64>>,
    /* private fields */
}

Implementation of a standard Naive Bayes classifier.

This classifier uses Laplace smoothing, the degree of which can be controlled with the alpha parameter.

Parameters

• alpha: The Laplace smoothing factor.
• probability_of_class: HashMap storing the probabilities of each class.
• probability_of_feat_by_class: HashMap storing the probabilities of each feature given a class.
• state: PhantomData indicating whether the classifier has been fit.

Type parameters

• State: Indicates whether the classifier has been fit. Can either be Fit or Unfit.

Example

use ducky_learn::naive_bayes::StdNaiveBayes;

// Define train and test data
let x_train: Vec<Vec<f64>> = vec![
    vec![1.0, 2.0, 3.0],
    vec![2.0, 3.0, 4.0],
    vec![3.0, 4.0, 5.0],
];
let y_train: Vec<String> = vec!["class1".to_string(), "class2".to_string(), "class1".to_string()];

let x_test: Vec<Vec<f64>> = vec![
    vec![1.5, 2.5, 3.5],
    vec![2.5, 3.5, 4.5],
];

let mut nb = StdNaiveBayes::new(1.0);
let nb = nb.fit(&x_train, &y_train);
let y_pred = nb.predict(&x_test);

// y_pred will hold the predicted classes for x_test

Fields

alpha: f64

probability_of_class: HashMap<String, f64>

probability_of_feat_by_class: HashMap<String, HashMap<String, f64>>

Implementations

impl StdNaiveBayes

pub fn new(alpha: f64) -> Self

Constructs a new, unfitted StdNaiveBayes classifier with a specified alpha value.

Parameters

• alpha: The Laplace smoothing factor.

Returns

A new StdNaiveBayes instance.

pub fn fit(self, x: &Vec<Vec<f64>>, y: &Vec<String>) -> StdNaiveBayes<Fit>

Fits the StdNaiveBayes classifier to the training data.

Parameters

• x: The training data.
• y: The target values.

Returns

The fitted StdNaiveBayes classifier.

impl StdNaiveBayes<Fit>

pub fn predict(&self, x: &Vec<Vec<f64>>) -> Vec<String>

Predicts the target values for the given data.

Parameters

• x: The data to predict target values for.

Returns

The predicted target values.

Panics

This function will panic if the classifier has not been fit.

Trait Implementations

impl<State: Debug> Debug for StdNaiveBayes<State>

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

Formats the value using the given formatter.