Struct OutlierTransformer 

pub struct OutlierTransformer<State = Untrained> { /* private fields */ }

Outlier transformer for handling extreme values through transformation

Implementations

impl OutlierTransformer<Untrained>

pub fn new() -> Self

Create a new OutlierTransformer with default configuration

pub fn log() -> Self

Create a log transformation for outliers

pub fn log1p() -> Self

Create a log1p transformation for outliers

pub fn sqrt() -> Self

Create a square root transformation for outliers

pub fn box_cox() -> Self

Create a Box-Cox transformation with automatic lambda

pub fn box_cox_fixed(lambda: Float) -> Self

Create a Box-Cox transformation with fixed lambda

pub fn quantile_uniform(n_quantiles: usize) -> Self

Create a quantile transformation to uniform distribution

pub fn quantile_normal(n_quantiles: usize) -> Self

Create a quantile transformation to normal distribution

pub fn robust_scale() -> Self

Create a robust scaling transformation

pub fn interpolate(threshold: Float, detection_method: &str) -> Self

Create an interpolation transformation

pub fn smooth(window_size: usize, threshold: Float) -> Self

Create a smoothing transformation

pub fn trim(lower_percentile: Float, upper_percentile: Float) -> Self

Create a trimming transformation

pub fn method(self, method: OutlierTransformationMethod) -> Self

Set the transformation method

pub fn outlier_threshold(self, threshold: Float) -> Self

Set the outlier detection threshold

pub fn detection_method(self, method: String) -> Self

Set the outlier detection method

pub fn lower_percentile(self, percentile: Float) -> Self

Set the lower percentile for trimming

pub fn upper_percentile(self, percentile: Float) -> Self

Set the upper percentile for trimming

pub fn smoothing_window(self, window: usize) -> Self

Set the smoothing window size

pub fn n_quantiles(self, n_quantiles: usize) -> Self

Set the number of quantiles

pub fn handle_negatives(self, handle: bool) -> Self

Set whether to handle negative values

pub fn log_epsilon(self, epsilon: Float) -> Self

Set the epsilon for log transformations

pub fn feature_wise(self, feature_wise: bool) -> Self

Set whether to apply transformations feature-wise

impl OutlierTransformer<Trained>

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

Get the number of features seen during fit

pub fn transformation_params(&self) -> Option<&TransformationParameters>

Get transformation parameters

pub fn feature_stats( &self, feature_idx: usize, ) -> Option<&FeatureTransformationParams>

Get transformation statistics for a specific feature

Trait Implementations

impl<State: Clone> Clone for OutlierTransformer<State>

fn clone(&self) -> OutlierTransformer<State>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<State: Debug> Debug for OutlierTransformer<State>

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

Formats the value using the given formatter.

impl Default for OutlierTransformer<Untrained>

fn default() -> Self

Returns the “default value” for a type.

impl Fit<ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>, ()> for OutlierTransformer<Untrained>

type Fitted = OutlierTransformer<Trained>

The fitted model type

fn fit(self, x: &Array2<Float>, _y: &()) -> Result<Self::Fitted>

Fit the model to the provided data with validation

fn fit_with_validation( self, x: &X, y: &Y, _x_val: Option<&X>, _y_val: Option<&Y>, ) -> Result<(Self::Fitted, FitMetrics), SklearsError>

where Self: Sized,

Fit with custom validation and early stopping

impl Transform<ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>> for OutlierTransformer<Trained>

fn transform(&self, x: &Array2<Float>) -> Result<Array2<Float>>

Transform the input data