Struct rusty_machine::learning::toolkit::kernel::KernelProd

pub struct KernelProd<T, U>where
    T: Kernel,
    U: Kernel,{ /* private fields */ }

The pointwise product of two kernels

This struct should not be directly instantiated but instead is created when we multiply two kernels together.

Note that it will be more efficient to implement the final kernel manually yourself. However this provides an easy mechanism to test different combinations.

Examples

use rusty_machine::learning::toolkit::kernel::{Kernel, Polynomial, HyperTan, KernelArith};

let poly_ker = Polynomial::new(1f64,2f64,3f64);
let hypert_ker = HyperTan::new(1f64,2.5);

let poly_plus_hypert_ker = KernelArith(poly_ker) * KernelArith(hypert_ker);

println!("{0}", poly_plus_hypert_ker.kernel(&[1f64,2f64,3f64],
                                            &[3f64,1f64,2f64]));

Trait Implementations

impl<T: Debug, U: Debug> Debug for KernelProd<T, U>where
    T: Kernel,
    U: Kernel,

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

Formats the value using the given formatter.

impl<T, U> Kernel for KernelProd<T, U>where
    T: Kernel,
    U: Kernel,

Computes the product of the two associated kernels.

fn kernel(&self, x1: &[f64], x2: &[f64]) -> f64

The kernel function.