Struct GridSpline2dState 

pub struct GridSpline2dState {
    pub coeffs: Vec<Vec<f64>>,
    pub log_lambda: f64,
    pub sigma2: Vec<f64>,
    pub restricted_loglik: f64,
    pub chol: Vec<f64>,
    pub axis_lo: [f64; 2],
    pub axis_h: [f64; 2],
    pub axis_cells: [u64; 2],
    pub m_axis: u64,
}

Serializable snapshot of a GridSpline2dFit (#1031 persistence prerequisite). The grid is deliberately NOT a formula fast path — it is an ANOVA pair component (#975 carve) — so there is no FitResult variant; this state is what the carve’s persistence payload serializes and what from_state replays for an exact predict.

Predict needs the MEAN (coeffs + the 16-entry tensor basis row, which is a pure function of axes/m_axis) and the VARIANCE (σ²·x'(X'WX+λS)⁻¹x through the retained Cholesky factor chol). All of that — and nothing about the training rows — lives on the fit already, so the state is a verbatim snapshot: no design CSR, no re-factor on load.

Fields

coeffs: Vec<Vec<f64>>

Per response dimension: row-major coefficients g = j1·(K+3) + j2.

log_lambda: f64

sigma2: Vec<f64>

Per response dimension: profiled (or supplied) observation variance σ².

restricted_loglik: f64

chol: Vec<f64>

Lower Cholesky factor of X'WX + λS (unit-σ² scale), p × p row-major — the factored posterior precision the variance term solves against.

axis_lo: [f64; 2]

Per axis lower corner of the basis bounding box.

axis_h: [f64; 2]

Per axis cell width h = (hi − lo)/K.

axis_cells: [u64; 2]

Per axis cell count K.

m_axis: u64

Basis count per axis, K + 3 (so p = m_axis²).

Trait Implementations

impl Clone for GridSpline2dState

fn clone(&self) -> GridSpline2dState

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GridSpline2dState

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for GridSpline2dState

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for GridSpline2dState

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.