Struct TransmissionKLBackgroundModel 

pub struct TransmissionKLBackgroundModel<'a> {
    pub inner: &'a dyn FitModel,
    pub inv_sqrt_energies: Vec<f64>,
    pub b0_index: usize,
    pub b1_index: usize,
    pub n_params: usize,
}

KL-compatible background model for transmission data.

Given a transmission model T_inner(θ), predicts:

T_out(E) = T_inner(E) + b₀ + b₁/√E

where b₀ and b₁ are the additive background parameters at indices b0_index and b1_index in the parameter vector.

Unlike NormalizedTransmissionModel (which uses Anorm * T + BackA + BackB/√E + BackC√E with 4 free parameters), this model:

• Has only 2 background parameters (b₀, b₁), reducing overfitting risk
• Constrains b₀, b₁ ≥ 0 via parameter bounds (physical: background adds counts, never subtracts), ensuring T_out > 0 for valid Poisson NLL
• Does NOT multiply T_inner by a normalization factor — normalization is handled separately (nuisance estimation for counts, or pre-processing for transmission data)

Gradient

• ∂T_out/∂nₖ = ∂T_inner/∂nₖ = -σₖ(E)·T_inner(E) (same as bare model)
• ∂T_out/∂b₀ = 1
• ∂T_out/∂b₁ = 1/√E

Fields

inner: &'a dyn FitModel

Underlying transmission model (density parameters only).

inv_sqrt_energies: Vec<f64>

Precomputed 1/√E for each energy bin.

b0_index: usize

Index of b₀ (constant background) in the parameter vector.

b1_index: usize

Index of b₁ (1/√E background) in the parameter vector.

n_params: usize

Total parameter count in the wrapped model.

Trait Implementations

impl<'a> FitModel for TransmissionKLBackgroundModel<'a>

fn evaluate(&self, params: &[f64]) -> Result<Vec<f64>, FittingError>

Evaluate the model for the given parameters.

fn analytical_jacobian( &self, params: &[f64], free_param_indices: &[usize], y_current: &[f64], ) -> Option<FlatMatrix>

Optionally provide an analytical Jacobian.

impl<'a> ForwardModel for TransmissionKLBackgroundModel<'a>

fn predict(&self, params: &[f64]) -> Result<Vec<f64>, FittingError>

Predict model output for the given parameter vector.

fn n_data(&self) -> usize

Number of data points in the model output.

fn n_params(&self) -> usize

Number of parameters (total, including fixed).

fn jacobian( &self, _params: &[f64], _free_param_indices: &[usize], _y_current: &[f64], ) -> Option<Vec<Vec<f64>>>

Analytical Jacobian (column-major layout).