Struct RandomFourierFeatures 

pub struct RandomFourierFeatures { /* private fields */ }

Random Fourier Features for scalable kernel approximation.

This struct stores the random frequencies sampled from the spectral distribution of the kernel, allowing efficient feature computation.

Implementations

impl RandomFourierFeatures

pub fn new(input_dim: usize, config: RffConfig) -> Result<Self>

Create a new Random Fourier Features transformer.

Arguments

• input_dim - Dimension of input vectors
• config - RFF configuration

Example

use tensorlogic_sklears_kernels::random_features::{
    RandomFourierFeatures, RffConfig, KernelType
};

let config = RffConfig::new(KernelType::Rbf { gamma: 1.0 }, 50);
let rff = RandomFourierFeatures::new(10, config).unwrap();

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

Transform a single input vector to the random feature space.

Arguments

• x - Input vector of dimension input_dim

Returns

Feature vector of dimension 2 * n_components

pub fn transform_batch(&self, data: &[Vec<f64>]) -> Result<Vec<Vec<f64>>>

Transform multiple input vectors to the random feature space.

Arguments

• data - List of input vectors

Returns

Matrix of feature vectors (one row per input)

pub fn approximate_kernel(&self, x: &[f64], y: &[f64]) -> Result<f64>

Approximate the kernel value between two vectors.

K(x, y) ≈ z(x)^T z(y)

pub fn approximate_kernel_matrix( &self, data: &[Vec<f64>], ) -> Result<Vec<Vec<f64>>>

Approximate the kernel matrix for a set of data points.

K[i,j] ≈ z(x_i)^T z(x_j)

pub fn input_dim(&self) -> usize

Get the input dimension.

pub fn output_dim(&self) -> usize

Get the output dimension (feature space dimension).

pub fn n_components(&self) -> usize

Get the number of random components.

pub fn kernel_type(&self) -> &KernelType

Get the kernel type being approximated.

pub fn random_weights(&self) -> &[Vec<f64>]

Get the random weights matrix.

Trait Implementations

impl Clone for RandomFourierFeatures

fn clone(&self) -> RandomFourierFeatures

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RandomFourierFeatures

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

Formats the value using the given formatter.