pod

pod(
    poi_col,
    threshold,
    nuisance_col=None,
    slice_values=None,
    bandwidth_ratio=None,
    n_boot=1000,
    model_override='auto',
    force_degree=None,
    n_jobs=None,
    nuisance_dists=None,
    progress_callback=None,
)

Runs the generalized Probability of Detection (PoD) analysis.

Parameters

Name	Type	Description	Default
poi_col	list | str	The parameter(s) of interest (e.g., ‘Crack Length’, [‘Angle’, ‘Depth’]).	required
threshold	float	The failure threshold (e.g., 4.0 dB).	required
nuisance_col	list | str | None	The nuisance parameters to marginalize over via MC integration.	None
slice_values	dict | None	The sliced parameters and their values.	None
bandwidth_ratio	float | None	Bandwidth of the variance smoother. None (default) chooses it automatically by leave-one-out cross-validation; a number fixes it as that fraction of the (standardised) input range.	None
n_boot	int	Bootstrap iterations for confidence bounds.	1000
model_override	str	Force a model type. One of “auto”, “polynomial”, or “kriging”. Defaults to “auto”.	'auto'
force_degree	int | None	When model_override=“polynomial”, use this degree. Defaults to None (CV selects).	None
n_jobs	int | None	Number of worker processes for the bootstrap. None or 1 (default) runs on a single core; -1 uses all cores except two, so the machine stays responsive; n uses n.	None

Returns

Name	Type	Description
Dict	Dict[str, Any]	Dictionary containing models, curves, and fit statistics.

Examples

results = study.pod(poi_col="Length", threshold=0.5)
print(results['dist_info'])