Struct SparsePolynomialFeatures 

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

SparsePolynomialFeatures generates sparse polynomial and interaction features

This is a memory-efficient version of PolynomialFeatures designed for high-dimensional data. Instead of storing all possible polynomial combinations, it stores only the terms that would actually be non-zero, using a sparse representation.

This is particularly useful when:

• The input dimensionality is high (hundreds or thousands of features)
• Many input features are zero or near-zero
• Memory usage is a concern
• You want to limit the total number of generated features

Implementations

impl SparsePolynomialFeatures<Untrained>

pub fn new() -> Self

Create a new SparsePolynomialFeatures

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

Set the degree of polynomial features

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

Set whether to include interaction terms only

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

Set whether to include bias term

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

Set maximum number of terms to generate

pub fn min_coefficient(self, min_coefficient: Float) -> Self

Set minimum coefficient magnitude threshold

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

Set whether to sort terms

impl SparsePolynomialFeatures<Trained>

pub fn n_features_in(&self) -> usize

Get the number of input features

pub fn n_output_features(&self) -> usize

Get the number of output features

pub fn sparse_terms(&self) -> &[SparseCoefficient]

Get the sparse terms

pub fn memory_info(&self) -> SparseMemoryInfo

Get memory usage information

Trait Implementations

impl Default for SparsePolynomialFeatures<Untrained>

fn default() -> Self

Returns the “default value” for a type.

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

type Fitted = SparsePolynomialFeatures<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 SparsePolynomialFeatures<Trained>

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

Transform the input data