Module kernel 

Kernel methods (Kernel PCA, Kernel Ridge Regression, kernel functions) Kernel Methods

This module provides kernel-based algorithms for nonlinear machine learning:

• Kernel Functions (kernels): A library of kernel functions (Linear, Polynomial, RBF/Gaussian, Laplacian, Sigmoid), Gram matrix computation, and kernel centering.

• Kernel PCA (kpca): Nonlinear dimensionality reduction via the kernel trick, with pre-image estimation and automatic parameter selection.

• Kernel Ridge Regression (kernel_ridge): Tikhonov-regularized regression in kernel space, with closed-form LOO-CV and multi-output support.

Re-exports

pub use kernel_ridge::KernelRidgeRegression;

pub use kernels::center_kernel_matrix;

pub use kernels::center_kernel_matrix_test;

pub use kernels::cross_gram_matrix;

pub use kernels::estimate_rbf_gamma;

pub use kernels::gram_matrix;

pub use kernels::is_positive_semidefinite;

pub use kernels::kernel_alignment;

pub use kernels::kernel_diagonal;

pub use kernels::kernel_eval;

pub use kernels::KernelType;

pub use kpca::KernelPCA;

Modules

kernel_ridge

Kernel Ridge Regression Kernel Ridge Regression

kernels

Kernel functions library (Linear, Polynomial, RBF, Laplacian, Sigmoid) Kernel Functions Library

kpca

Kernel PCA for nonlinear dimensionality reduction Kernel PCA (Principal Component Analysis)