Struct Rvm 

pub struct Rvm { /* private fields */ }

A fitted kernel Relevance Vector Machine.

Implementations

impl Rvm

pub fn fit( x: &[f64], n: usize, d: usize, t: &[f64], config: RvmConfig, ) -> CsResult<Self>

Fit a kernel RVM to n training points x ∈ ℝ^{n×d} (row-major) and targets t.

Errors

• CsError::InvalidParameter for n == 0, d == 0, or a non-positive RBF length-scale.
• CsError::ShapeMismatch / CsError::DimensionMismatch on bad shapes.

pub fn n_relevance(&self) -> usize

Number of relevance vectors retained (excludes the bias term).

pub fn noise_std(&self) -> f64

Estimated noise standard deviation σ = 1/√β.

pub fn predict_one(&self, query: &[f64]) -> CsResult<f64>

Predict the target at a single query input x⋆ ∈ ℝ^d.

Errors

CsError::DimensionMismatch if query.len() != d.

pub fn predict(&self, queries: &[f64], n_q: usize) -> CsResult<Vec<f64>>

Predict targets for n_q query points x ∈ ℝ^{n_q×d} (row-major).

Errors

CsError::ShapeMismatch if queries.len() != n_q * d.

Trait Implementations

impl Clone for Rvm

fn clone(&self) -> Rvm

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Rvm

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

Formats the value using the given formatter.