Struct GaussianVectorLikelihood 

pub struct GaussianVectorLikelihood {
    pub precision: Array1<f64>,
    pub factor: Option<Array2<f64>>,
    pub row_weights: Option<Array1<f64>>,
}

Gaussian vector likelihood with identity link.

log p(Y|η) = −½ Σ_n w_n · rᵀ W r where r = Y_n − η_n and W is the per-output precision matrix. For Isotropic / Diagonal W = diag(prec); for LowRank it is W = diag(prec) + F · Fᵀ, with F carried alongside the diagonal here.

(Up to the constant log-determinant of the noise covariance, dropped here because it does not depend on β or the latent t; the determinant is accounted for in the REML score, not the inner likelihood.)

Fields

precision: Array1<f64>

Per-output diagonal precision (length M). For Isotropic / Diagonal / LowRank this is the diagonal piece of the precision matrix (1/σ_m² for Diagonal/Isotropic; diag for LowRank).

factor: Option<Array2<f64>>

Optional dense rank-r factor F of size (M, r) such that the full per-row precision is diag(precision) + F · Fᵀ. None for the Isotropic / Diagonal cases.

row_weights: Option<Array1<f64>>

Optional row weights (length N), or None for uniform.

Implementations

impl GaussianVectorLikelihood

pub fn from_target( target: &VectorResponseTarget, ) -> Result<Self, EstimationError>

Trait Implementations

impl Clone for GaussianVectorLikelihood

fn clone(&self) -> GaussianVectorLikelihood

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GaussianVectorLikelihood

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

Formats the value using the given formatter.

impl VectorLikelihood for GaussianVectorLikelihood

fn log_lik(&self, eta: ArrayView2<'_, f64>, y: ArrayView2<'_, f64>) -> f64

log p(Y | η).

fn grad_eta( &self, eta: ArrayView2<'_, f64>, y: ArrayView2<'_, f64>, ) -> Array2<f64>

∂ log p(Y | η) / ∂ η, shape (N, M).

fn hess_diag( &self, eta: ArrayView2<'_, f64>, y: ArrayView2<'_, f64>, ) -> Array2<f64>

Diagonal of the per-row Hessian −∂² log p / ∂ η ∂ η, shape (N, M). This is the per-row block consumed by solver/arrow_schur.rs.

fn hess_block( &self, eta: ArrayView2<'_, f64>, y: ArrayView2<'_, f64>, ) -> Array3<f64>

Per-row dense Hessian block −∂² log p / ∂η_a ∂η_b, shape (N, M, M).