Struct PolynomialFeatures 

pub struct PolynomialFeatures { /* private fields */ }

Degree-2 polynomial feature generator.

Generates original features plus all degree-2 combinations (squared terms and pairwise interactions). When interaction_only is constructed, squared terms (x_i * x_i) are excluded, leaving only cross-interactions (x_i * x_j where i < j).

Because the transform is purely deterministic and stateless, both update_and_transform and transform behave identically, and reset clears only the cached input dimensionality.

Implementations

impl PolynomialFeatures

pub fn new() -> Self

Create a polynomial feature generator with full degree-2 expansion.

Includes original features, squared terms, and pairwise interactions.

use irithyll::preprocessing::PolynomialFeatures;
use irithyll::pipeline::StreamingPreprocessor;

let poly = PolynomialFeatures::new();
let out = poly.transform(&[2.0, 3.0]);
// [a, b, a², ab, b²] = [2, 3, 4, 6, 9]
assert_eq!(out, vec![2.0, 3.0, 4.0, 6.0, 9.0]);

pub fn interaction_only() -> Self

Create a polynomial feature generator that only includes cross-interactions.

Squared terms (x_i * x_i) are excluded; only x_i * x_j where i < j are generated.

use irithyll::preprocessing::PolynomialFeatures;
use irithyll::pipeline::StreamingPreprocessor;

let poly = PolynomialFeatures::interaction_only();
let out = poly.transform(&[2.0, 3.0]);
// [a, b, ab] = [2, 3, 6]
assert_eq!(out, vec![2.0, 3.0, 6.0]);

pub fn is_interaction_only(&self) -> bool

Returns true if this generator excludes squared terms.

Trait Implementations

impl Clone for PolynomialFeatures

fn clone(&self) -> PolynomialFeatures

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PolynomialFeatures

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

Formats the value using the given formatter.

impl Default for PolynomialFeatures

fn default() -> Self

Returns the “default value” for a type.

impl StreamingPreprocessor for PolynomialFeatures

fn update_and_transform(&mut self, features: &[f64]) -> Vec<f64>

Update internal statistics from this sample and return transformed features.

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

Transform features using current statistics without updating them.

fn output_dim(&self) -> Option<usize>

Number of output features, or None if unknown until the first sample.

fn reset(&mut self)

Reset to initial (untrained) state.