Struct MinMaxScaler 

pub struct MinMaxScaler { /* private fields */ }

Online min-max scaler that scales features to a configurable range.

When a feature has not been observed, transform returns the original value. When a feature is constant (min == max), the output is the midpoint of the target range to avoid NaN.

Implementations

impl MinMaxScaler

pub fn new(feature_count: usize) -> Result<Self, RillError>

Create a new min-max scaler scaling to [0, 1].

pub fn with_range( feature_count: usize, target_min: f64, target_max: f64, ) -> Result<Self, RillError>

Create a new min-max scaler scaling to [target_min, target_max].

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

The per-feature minima (None if not yet observed).

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

The per-feature maxima (None if not yet observed).

Trait Implementations

impl Clone for MinMaxScaler

fn clone(&self) -> MinMaxScaler

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MinMaxScaler

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

Formats the value using the given formatter.

impl Transformer for MinMaxScaler

fn input_dim(&self) -> usize

Expected number of input features.

fn output_dim(&self) -> usize

Number of features produced by transform.

fn transform(&self, features: &[f64]) -> Result<Vec<f64>, RillError>

Transform features using the current internal state.

fn update(&mut self, features: &[f64]) -> Result<(), RillError>

Update internal statistics using raw features.

fn samples_seen(&self) -> u64

How many samples the transformer has seen.

fn reset(&mut self)

Reset the transformer to its initial state.