Struct StructuredKernelInterpolationGPR 

pub struct StructuredKernelInterpolationGPR {
    pub kernel: Box<dyn Kernel>,
    pub grid_size: usize,
    pub interpolation_method: InterpolationMethod,
    pub grid_bounds_method: GridBoundsMethod,
    pub boundary_extension: f64,
    pub noise_variance: f64,
    pub use_toeplitz: bool,
    pub random_state: Option<u64>,
    pub cg_tolerance: f64,
    pub max_cg_iterations: usize,
}

Structured Kernel Interpolation Gaussian Process Regressor

Uses structured interpolation on regular grids to achieve scalable GP inference with O(n log n) computational complexity for large datasets.

Example

use sklears_gaussian_process::{StructuredKernelInterpolationGPR, InterpolationMethod, kernels::RBF};
use sklears_core::prelude::*;
// SciRS2 Policy - Use scirs2-autograd for ndarray types and operations

let kernel = Box::new(RBF::new(1.0));
let model = StructuredKernelInterpolationGPR::new(kernel)
    .grid_size(64)
    .interpolation_method(InterpolationMethod::Linear);

let X = Array2::from_shape_vec((100, 2), (0..200).map(|x| x as f64).collect()).unwrap();
let y = Array1::from_vec((0..100).map(|x| (x as f64).sin()).collect());

let trained_model = model.fit(&X.view(), &y.view()).unwrap();
let predictions = trained_model.predict(&X.view()).unwrap();

Fields

kernel: Box<dyn Kernel>

Base kernel to interpolate

grid_size: usize

Grid size for each dimension

interpolation_method: InterpolationMethod

Interpolation method

grid_bounds_method: GridBoundsMethod

Grid bounds method

boundary_extension: f64

Boundary extension factor

noise_variance: f64

Noise variance parameter

use_toeplitz: bool

Whether to use Toeplitz structure for regular grids

random_state: Option<u64>

Random state for reproducible results

cg_tolerance: f64

Tolerance for conjugate gradient solver

max_cg_iterations: usize

Maximum iterations for conjugate gradient

Implementations

impl StructuredKernelInterpolationGPR

pub fn new(kernel: Box<dyn Kernel>) -> Self

Create a new SKI Gaussian process regressor

pub fn grid_size(self, size: usize) -> Self

Set the grid size

pub fn interpolation_method(self, method: InterpolationMethod) -> Self

Set the interpolation method

pub fn grid_bounds_method(self, method: GridBoundsMethod) -> Self

Set the grid bounds method

pub fn boundary_extension(self, extension: f64) -> Self

Set boundary extension factor

pub fn noise_variance(self, variance: f64) -> Self

Set noise variance

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

Set whether to use Toeplitz structure

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

Set random state

pub fn compute_approximation_info( &self, X: &ArrayView2<'_, f64>, grid_points: &[Array1<f64>], ) -> SklResult<SkiApproximationInfo>

Compute approximation quality metrics

Trait Implementations

impl Clone for StructuredKernelInterpolationGPR

fn clone(&self) -> StructuredKernelInterpolationGPR

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for StructuredKernelInterpolationGPR

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

Formats the value using the given formatter.

impl Default for StructuredKernelInterpolationGPR

fn default() -> Self

Returns the “default value” for a type.

impl Estimator for StructuredKernelInterpolationGPR

type Config = StructuredKernelInterpolationGPR

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<ViewRepr<&f64>, Dim<[usize; 2]>>, ArrayBase<ViewRepr<&f64>, Dim<[usize; 1]>>, SkiGprTrained> for StructuredKernelInterpolationGPR

type Fitted = SkiGprTrained

The fitted model type

fn fit( self, X: &ArrayView2<'_, f64>, y: &ArrayView1<'_, f64>, ) -> SklResult<SkiGprTrained>

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