Struct TruingUncertaintyDiagnosticsV1 

pub struct TruingUncertaintyDiagnosticsV1 {

    pub chi_square: f64,
    pub prior_penalty: f64,
    pub penalized_chi_square: f64,
    pub effective_parameter_count: Option<f64>,
    pub effective_degrees_of_freedom: Option<f64>,
    pub reduced_chi_square: Option<f64>,
    pub bc_sensitivity_ratio: f64,
    pub data_condition_number: f64,
    pub map_scaled_gradient_inf_norm: f64,
    pub map_convergence_criterion: Option<TruingMapConvergenceCriterionV1>,
    pub map_objective_poll_radius: Option<f64>,
    pub map_max_objective_poll_improvement: Option<f64>,
    pub map_objective_poll_evaluations: usize,
}

Fit and approximation diagnostics.

Fields

chi_square: f64

Data chi-square, sum((prediction - observation) / sigma)^2.

prior_penalty: f64

Explicit prior contribution to the penalized chi-square.

penalized_chi_square: f64

chi_square + prior_penalty, minimized by the MAP fitter.

effective_parameter_count: Option<f64>

Effective number of parameters learned from the observations, trace(I_data * I_posterior^-1).

effective_degrees_of_freedom: Option<f64>

Effective residual degrees of freedom, n - effective_parameter_count.

reduced_chi_square: Option<f64>

Data chi-square divided by effective residual degrees of freedom.

bc_sensitivity_ratio: f64

Fractional BC sensitivity relative to fractional MV sensitivity.

data_condition_number: f64

Collinearity condition diagnostic for the weighted data Jacobian.

map_scaled_gradient_inf_norm: f64

Infinity norm of the penalized-chi-square half-gradient in scaled coordinates at the reported MAP. This is an informative local-model diagnostic. It is the primary convergence test, but the deliberately broad finite-difference stencil can disagree with the trajectory solver’s much finer numerical surface. In that case TruingMapConvergenceCriterionV1::ObjectiveMesh records a direct- objective verification instead.

map_convergence_criterion: Option<TruingMapConvergenceCriterionV1>

Criterion that verified the reported MAP, or None if the optimizer did not converge.

map_objective_poll_radius: Option<f64>

Final scaled pattern-poll radius for objective-mesh convergence. 100 fps and 0.1 BC are one scaled unit. None when the gradient criterion was used or the poll did not reach its resolution target.

map_max_objective_poll_improvement: Option<f64>

Largest direct chi-square improvement found in the final pattern poll. Objective-mesh convergence requires this to be no greater than 1e-8 in absolute penalized chi-square units.

map_objective_poll_evaluations: usize

Number of direct objective evaluations used by the fallback poll.

Trait Implementations

impl Clone for TruingUncertaintyDiagnosticsV1

fn clone(&self) -> TruingUncertaintyDiagnosticsV1

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TruingUncertaintyDiagnosticsV1

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

Formats the value using the given formatter.

impl PartialEq for TruingUncertaintyDiagnosticsV1

fn eq(&self, other: &TruingUncertaintyDiagnosticsV1) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for TruingUncertaintyDiagnosticsV1

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Copy for TruingUncertaintyDiagnosticsV1

impl StructuralPartialEq for TruingUncertaintyDiagnosticsV1