core.SimulationStudy

SimulationStudy()

A workflow manager for simulation reliability assessment.

Attributes

Name	Type	Description
inputs	List[str]	List of input variable names.
outcome	str	Name of the outcome variable.
data	pd.DataFrame	The raw simulation data.
clean_data	pd.DataFrame	Data that has passed validation.
sufficiency_results	pd.DataFrame	The latest diagnostic results.
pod_results	Dict	Results from the latest PoD analysis.
plots	Dict	Stores the latest generated figures.
linear_pod_results	Dict	Results from the classical linear analysis.
linear_plots	Dict	Stores figures from the classical linear analysis.
models_cache	Dict	Cached mean models to prevent redundant fitting.
variance_cache	Dict	Cached variance models (residuals/bandwidth).
pod_curves_cache	Dict	Cached integrated PoD curves.
threshold_spectrum_cache	Dict	Cached threshold spectrum.

Examples

from digiqual.core import SimulationStudy

# 1. Clean Initialization
study = SimulationStudy()

# 2. Add data (automatically infers input columns)
study.add_data(df, outcome_col='Signal')

# 3. Run diagnostics with custom strict thresholds
results = study.diagnose(max_gap_ratio=0.10, min_r2_score=0.75)

Methods

Name	Description
add_data	Ingests raw simulation data, configures columns, and manages the cache.
compute_pod_spectrum	Pre-calculates a spectrum of PoD curves across a range of signal thresholds.
diagnose	Runs statistical diagnostics to evaluate if the current sample size is sufficient.
estimate_compute_time	Physics-aware heuristic to estimate PoD computation time in seconds.
get_data_summary	Calculates the minimum, maximum, and median values for a given column.
get_unassigned_parameters	Calculates which input variables are not currently assigned as a Parameter
linear_pod	Runs the classical linear a-hat vs a PoD analysis with bootstrapped bounds.
optimise	Runs the automated Active Learning loop (Initialize -> Execute -> Diagnose -> Refine).
pod	Runs the generalized Probability of Detection (PoD) analysis.
refine	Identifies gaps or high-variance regions and suggests new simulation points.
update_slice	Updates the evaluated slice for the PoD surface using Layer 3 Caching.
visualise	Generates and displays diagnostic plots (Signal Model and PoD Curve).