Struct WOEEncoder 

pub struct WOEEncoder { /* private fields */ }

Weight of Evidence (WOE) Encoder for converting categorical features using target information

WOE encoding transforms categorical features based on the relationship between each category and a binary target variable. It’s particularly useful for credit scoring and other binary classification tasks.

WOE = ln(P(target=1|category) / P(target=0|category))

Implementations

impl WOEEncoder

pub fn new( regularization: f64, handleunknown: &str, unknownvalue: f64, ) -> Result<Self>

Creates a new WOEEncoder

Arguments

• regularization - Small value added to prevent division by zero (default: 0.5)
• handleunknown - How to handle unknown categories (‘error’, ‘global_woe’, or ‘use_encoded_value’)
• unknownvalue - Value to use for unknown categories (when handleunknown=“use_encoded_value”)

Returns

• Result<WOEEncoder> - The new encoder instance

pub fn with_defaults() -> Self

Creates a WOEEncoder with default settings

pub fn with_regularization(regularization: f64) -> Result<Self>

Creates a WOEEncoder with custom regularization

pub fn fit<S>(&mut self, x: &ArrayBase<S, Ix2>, y: &[f64]) -> Result<()>

where S: Data, S::Elem: Float + NumCast,

Fits the WOEEncoder to the input data

Arguments

• x - The input categorical data, shape (n_samples, n_features)
• y - The binary target values (0 or 1), length n_samples

Returns

• Result<()> - Ok if successful, Err otherwise

pub fn transform<S>(&self, x: &ArrayBase<S, Ix2>) -> Result<Array2<f64>>

where S: Data, S::Elem: Float + NumCast,

Transforms the input data using the fitted WOEEncoder

Arguments

• x - The input categorical data, shape (n_samples, n_features)

Returns

• Result<Array2<f64>> - The WOE-encoded data

pub fn fit_transform<S>( &mut self, x: &ArrayBase<S, Ix2>, y: &[f64], ) -> Result<Array2<f64>>

where S: Data, S::Elem: Float + NumCast,

Fits the encoder and transforms the data in one step

Arguments

• x - The input categorical data
• y - The binary target values

Returns

• Result<Array2<f64>> - The WOE-encoded data

pub fn is_fitted(&self) -> bool

Returns whether the encoder has been fitted

pub fn woe_maps(&self) -> Option<&Vec<HashMap<u64, f64>>>

Returns the learned WOE mappings if fitted

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

Returns the Information Values for each feature if fitted

Information Value interpretation:

• < 0.02: Not useful for prediction
• 0.02 - 0.1: Weak predictive power
• 0.1 - 0.3: Medium predictive power
• 0.3 - 0.5: Strong predictive power

• 0.5: Suspicious, too good to be true

pub fn global_woe(&self) -> f64

Returns the global WOE value (overall log-odds)

pub fn feature_importance_ranking(&self) -> Option<Vec<(usize, f64)>>

Returns features ranked by Information Value (descending order)

Returns

• Option<Vec<(usize, f64)>> - Vector of (feature_index, information_value) pairs

Trait Implementations

impl Clone for WOEEncoder

fn clone(&self) -> WOEEncoder

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for WOEEncoder

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

Formats the value using the given formatter.