Enum CoefficientPriorMean 

pub enum CoefficientPriorMean {
    Zero,
    Scalar(f64),
    Constant(Array1<f64>),
    Functional {
        metadata: Array1<f64>,
        evaluator: Arc<dyn Fn(&Array1<f64>) -> Array1<f64> + Send + Sync>,
    },
    KernelBasis {
        covariates: Array1<f64>,
        amplitude: f64,
        kernel: Arc<dyn Fn(&Array1<f64>) -> Array1<f64> + Send + Sync>,
    },
}

Programmatic prior mean for a coefficient penalty block.

The mean is evaluated once during penalty canonicalization and then enters the solver as the centering vector in (beta - mean)' S (beta - mean).

Variants

Zero

Scalar(f64)

Constant(Array1<f64>)

Functional

Fields

metadata: Array1<f64>

evaluator: Arc<dyn Fn(&Array1<f64>) -> Array1<f64> + Send + Sync>

KernelBasis

Covariate-functional mean mu(a) = amplitude * K(a) for a coefficient block.

Formula-level coefficient groups pass their row/covariate metadata as covariates; the user-supplied kernel returns the block-sized basis vector K(a) and the scalar amplitude supplies alpha.

Fields

covariates: Array1<f64>

amplitude: f64

kernel: Arc<dyn Fn(&Array1<f64>) -> Array1<f64> + Send + Sync>

Implementations

impl CoefficientPriorMean

pub const fn scalar(value: f64) -> Self

pub fn constant(values: Array1<f64>) -> Self

pub fn functional( metadata: Array1<f64>, evaluator: Arc<dyn Fn(&Array1<f64>) -> Array1<f64> + Send + Sync>, ) -> Self

pub fn kernel_basis( covariates: Array1<f64>, amplitude: f64, kernel: Arc<dyn Fn(&Array1<f64>) -> Array1<f64> + Send + Sync>, ) -> Self

pub fn evaluate( &self, block_dim: usize, context: &str, ) -> Result<Array1<f64>, PriorMeanError>

Trait Implementations

impl Clone for CoefficientPriorMean

fn clone(&self) -> CoefficientPriorMean

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for CoefficientPriorMean

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

Formats the value using the given formatter.

impl Default for CoefficientPriorMean

fn default() -> CoefficientPriorMean

Returns the “default value” for a type.