Struct FastfoodTransform 

pub struct FastfoodTransform<State = Untrained> {
    pub n_components: usize,
    pub gamma: Float,
    pub random_state: Option<u64>,
    /* private fields */
}

Fastfood Transform for efficient random Fourier features

The Fastfood transform uses structured matrices to approximate Gaussian random projections efficiently. It combines three structured transforms:

1. Random diagonal scaling (B)
2. Fast Walsh-Hadamard Transform (H)
3. Random permutation (Π)
4. Random diagonal scaling (G)

The overall transform is: G * H * Π * B * H This reduces complexity from O(d²) to O(d log d) while maintaining approximation quality for RBF kernels.

Parameters

• n_components - Number of random features to generate
• gamma - RBF kernel parameter (default: 1.0)
• random_state - Random seed for reproducibility

Examples

use sklears_kernel_approximation::fastfood::FastfoodTransform;
use sklears_core::traits::{Transform, Fit, Untrained}
use scirs2_core::ndarray::array;

let X = array![[1.0, 2.0, 3.0, 4.0], [2.0, 3.0, 4.0, 5.0]];

let fastfood = FastfoodTransform::new(8).gamma(0.5);
let fitted = fastfood.fit(&X, &()).unwrap();
let X_transformed = fitted.transform(&X).unwrap();
assert_eq!(X_transformed.shape(), &[2, 8]);

FastfoodTransform

Fields

n_components: usize

Number of random features

gamma: Float

RBF kernel gamma parameter

random_state: Option<u64>

Random seed

Implementations

impl FastfoodTransform<Untrained>

pub fn new(n_components: usize) -> Self

Create a new Fastfood transform

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

Set the gamma parameter for RBF kernel

pub fn random_state(self, seed: u64) -> Self

Set random state for reproducibility

Trait Implementations

impl<State: Clone> Clone for FastfoodTransform<State>

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

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<State: Debug> Debug for FastfoodTransform<State>

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

Formats the value using the given formatter.

impl Estimator for FastfoodTransform<Untrained>

type Config = ()

Configuration type for the estimator

type Error = SklearsError

Error type for the estimator

type Float = f64

The numeric type used by this estimator

fn config(&self) -> &Self::Config

Get estimator configuration

fn validate_config(&self) -> Result<(), SklearsError>

Validate estimator configuration with detailed error context

fn check_compatibility( &self, n_samples: usize, n_features: usize, ) -> Result<(), SklearsError>

Check if estimator is compatible with given data dimensions

fn metadata(&self) -> EstimatorMetadata

Get estimator metadata

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

type Fitted = FastfoodTransform<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 FastfoodTransform<Trained>

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

Transform the input data