Struct TimeBlockOperator 

pub struct TimeBlockOperator { /* private fields */ }

Row Jacobian operator for the survival time block. Channels (q0, q1, qd1) come from the three time designs; the g channel is zero.

Implementations

impl TimeBlockOperator

pub fn new(dq0: Array2<f64>, dq1: Array2<f64>, dqd1: Array2<f64>) -> Self

Trait Implementations

impl RowJacobianOperator for TimeBlockOperator

fn k(&self) -> usize

Dimension of the row primary state (survival: 4, Bernoulli: 1).

fn ncols(&self) -> usize

Number of coefficients in this block (= width of J_i).

fn nrows(&self) -> usize

Number of training rows.

fn apply_row(&self, row: usize, delta_beta: &[f64], out: &mut [f64])

Apply the row Jacobian: writes J_i · δβ ∈ R^K for row into out.

fn evaluate_full(&self) -> Array3<f64>

Materialise the full operator as an (n_rows × ncols × K) tensor.

fn scaled_design_by_sqrt_h(&self, h_full: &Array3<f64>) -> Array2<f64>

Build the sqrt(H)-scaled design W = stack_i sqrt(H_i) · J_i, flattened channel-major to (n_rows·K × ncols).

fn channel_flattened_column(&self, col: usize, out: &mut [f64])

Write the channel-flattened column col — the (n_rows · K) vector whose entry i·K + ch is J[i, col, ch] — into out.

fn channel_flattened_rows( &self, rows: Range<usize>, out: &mut ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>, )

Write channel-flattened rows for rows into out.