Model Validation

Given the centrality of auto- and cross-correlation functions in model validation, dymoval provides a class XCorrelation. Furthermore, it also provides a ValidationSession to cope with simulation results.

XCorrelation class

Constructor

XCorrelation(name, X, Y[, nlags, ...])

Cross-correlation of two signals X and Y.

Attributes

XCorrelation.name	XCorrelation object name.
XCorrelation.kind	Kind of the XCorrelation object.
XCorrelation.R	Auto- or cross-correlation array.

Methods

XCorrelation.estimate_whiteness([...])	Return the whiteness estimate based on the selected statistics.
XCorrelation.plot()	Plot the \(p imes q\) cross-correlation functions contained in R.

ValidationSession class

Constructor

ValidationSession(name, validation_dataset)

The ValidationSession class is used to validate models against a given dataset.

Attributes

ValidationSession.name	ValidationSession object name.
ValidationSession.dataset	The reference Dataset class object.
ValidationSession.outcome	Validation outcome.
ValidationSession.simulations_names	Names of the stored simulations.
ValidationSession.validation_thresholds	Input-residuals cross-correlation arrays.
ValidationSession.Ruu	Input auto-correlation arrays.
ValidationSession.Ree	Residuals auto-correlation arrays.
ValidationSession.Rue	Input-residuals cross-correlation arrays.

Methods

ValidationSession.append_simulation(...)	Append simulation results.
ValidationSession.drop_simulations(*sims)	Drop simulation results from the validation session object.
ValidationSession.plot_simulations([...])	Plot the stored simulation results.
ValidationSession.plot_residuals([...])	Plot the residuals auto- and cross-correlation functions.
ValidationSession.simulation_signals_list(...)	Return the signal name list of a given simulation.
ValidationSession.clear()	Remove all the stored simulation results in the current ValidationSession object.
ValidationSession.trim([tin, tout, verbosity])	Trim the Validation session ValidationSession object.

Functions

rsquared(x, y)	Return the \(R^2\) value of two signals.
compute_statistic(data[, statistic, weights])	Compute the statistic of a sequence of numbers.
whiteness_level(data[, data_bandwidths, ...])	Estimate the whiteness of the signal data.
validate_models(measured_in, measured_out, ...)	Validate models based on measured and simulated data.