Struct SpectralMixtureKernel 

pub struct SpectralMixtureKernel { /* private fields */ }

Spectral Mixture (SM) kernel.

K(x, y) = Σ_q w_q * exp(-2π² * Σ_d (x_d - y_d)² * v_q,d) * cos(2π * Σ_d (x_d - y_d) * μ_q,d)

where:

• w_q is the weight of component q
• μ_q,d is the mean frequency of component q in dimension d
• v_q,d is the variance of component q in dimension d

This kernel can approximate any stationary covariance function arbitrarily well as the number of components increases.

Implementations

impl SpectralMixtureKernel

pub fn new(components: Vec<SpectralComponent>) -> Result<Self>

Create a new Spectral Mixture kernel.

Arguments

• components - List of spectral components

Example

use tensorlogic_sklears_kernels::spectral_kernel::{SpectralMixtureKernel, SpectralComponent};
use tensorlogic_sklears_kernels::Kernel;

// Create a kernel with two periodic components
let components = vec![
    SpectralComponent::new_1d(1.0, 0.1, 0.01).unwrap(),  // Low frequency
    SpectralComponent::new_1d(0.5, 1.0, 0.1).unwrap(),   // High frequency
];
let kernel = SpectralMixtureKernel::new(components).unwrap();

pub fn new_1d(frequencies: Vec<(f64, f64, f64)>) -> Result<Self>

Create a simple 1D spectral mixture kernel with given frequencies.

Arguments

• frequencies - List of (weight, mean_frequency, variance) tuples

pub fn components(&self) -> &[SpectralComponent]

Get the components.

pub fn num_components(&self) -> usize

Get the number of components.

pub fn ndim(&self) -> usize

Get the number of dimensions.

Trait Implementations

impl Clone for SpectralMixtureKernel

fn clone(&self) -> SpectralMixtureKernel

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SpectralMixtureKernel

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

Formats the value using the given formatter.

impl Kernel for SpectralMixtureKernel

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

Compute kernel value between two inputs.

fn name(&self) -> &str

Get kernel name for identification.

fn compute_matrix(&self, inputs: &[Vec<f64>]) -> Result<Vec<Vec<f64>>>

Compute kernel matrix for a set of inputs.

fn is_psd(&self) -> bool

Check if kernel is positive semi-definite.