Struct RbfKernel 

pub struct RbfKernel<F> {
    pub gamma: Gamma<F>,
}

Radial Basis Function (Gaussian) kernel.

K(x, y) = exp(-gamma * ||x - y||^2)

Fields

gamma: Gamma<F>

The gamma parameter, a three-way Gamma<F> spec resolved at fit time (sklearn/svm/_base.py:235-243). Default Gamma::Scale (= 1 / (n_features * X.var())); Gamma::Auto = 1 / n_features; Gamma::Value is used verbatim.

Implementations

impl<F: Float> RbfKernel<F>

pub fn new() -> Self

Create a new RBF kernel with the default gamma='scale'.

pub fn with_gamma(gamma: F) -> Self

Create a new RBF kernel with an explicit float gamma (gamma=<float>, Gamma::Value).

pub fn with_gamma_scale() -> Self

Create a new RBF kernel with gamma='scale' (Gamma::Scale, sklearn’s default = 1 / (n_features * X.var())).

pub fn with_gamma_auto() -> Self

Create a new RBF kernel with gamma='auto' (Gamma::Auto = 1 / n_features).

Trait Implementations

impl<F: Clone> Clone for RbfKernel<F>

fn clone(&self) -> RbfKernel<F>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Copy> Copy for RbfKernel<F>

impl<F: Debug> Debug for RbfKernel<F>

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

Formats the value using the given formatter.

impl<F: Float> Default for RbfKernel<F>

fn default() -> Self

Returns the “default value” for a type.

impl<F: Float + Send + Sync> Kernel<F> for RbfKernel<F>

fn compute(&self, x: &[F], y: &[F]) -> F

Compute the kernel value between two vectors.

fn resolved_for_fit(&self, x: &Array2<F>) -> Self

Resolve any data-dependent kernel parameters against the training data at fit time, returning a copy of the kernel with those parameters fixed.

fn is_linear(&self) -> bool

Whether this is the linear kernel K(x, y) = x . y.