Struct ComplementNB 

pub struct ComplementNB<F> {
    pub alpha: F,
    pub class_prior: Option<Vec<F>>,
}

Complement Naive Bayes classifier.

A variant of Multinomial NB that uses complement-class statistics. More robust for imbalanced datasets.

Type Parameters

• F: The floating-point type (f32 or f64).

Fields

alpha: F

Additive (Laplace) smoothing parameter. Default: 1.0.

class_prior: Option<Vec<F>>

Optional user-supplied class priors. Note: ComplementNB does not use priors in the standard way (it uses complement weights), but this field is provided for API consistency with other NB variants.

Implementations

impl<F: Float> ComplementNB<F>

pub fn new() -> Self

Create a new ComplementNB with Laplace smoothing (alpha = 1.0).

pub fn with_alpha(self, alpha: F) -> Self

Set the Laplace smoothing parameter.

pub fn with_class_prior(self, priors: Vec<F>) -> Self

Set user-supplied class priors.

The priors must have length equal to the number of classes discovered during fitting. Note: ComplementNB uses complement-class weights rather than direct class priors, but the priors are stored for API consistency.

Trait Implementations

impl<F: Clone> Clone for ComplementNB<F>

fn clone(&self) -> ComplementNB<F>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for ComplementNB<F>

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

Formats the value using the given formatter.

impl<F: Float> Default for ComplementNB<F>

fn default() -> Self

Returns the “default value” for a type.

impl<F: Float + Send + Sync + 'static> Fit<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, ArrayBase<OwnedRepr<usize>, Dim<[usize; 1]>>> for ComplementNB<F>

fn fit( &self, x: &Array2<F>, y: &Array1<usize>, ) -> Result<FittedComplementNB<F>, FerroError>

Fit the Complement NB model.

Errors

• FerroError::ShapeMismatch if x and y have different numbers of rows.
• FerroError::InsufficientSamples if there are no samples.
• FerroError::InvalidParameter if any feature value is negative.

type Fitted = FittedComplementNB<F>

The fitted model type returned by fit.

type Error = FerroError

The error type returned by fit.

impl<F: Float + ToPrimitive + FromPrimitive + Send + Sync + 'static> PipelineEstimator<F> for ComplementNB<F>

fn fit_pipeline( &self, x: &Array2<F>, y: &Array1<F>, ) -> Result<Box<dyn FittedPipelineEstimator<F>>, FerroError>

Fit this estimator on the given data.