Settings

This section will cover the UQPCE settings.

Options

Option	Description
-i, --input-file	file containing variables (default: ‘input.yaml’)
-m, --matrix-file	file containing matrix elements (default: ‘run_matrix.dat’)
-r, --results-file	file containing results (default: ‘results.dat’)
--verification-matrix-file	file containing verification matrix elements (default: ‘verification_run_matrix.dat’)
--verification-results-file	file containing verification results (default: ‘verification_results.dat’)
--output-directory	directory that the outputs will be put in (default: ‘outputs’)
-c, --case	case of input data (default: None)
-s, --significance	significance level of the confidence interval (default: 0.05)
-o, --order	order of polynomial chaos expansion (default: 2)
-f, --user-func	allows the user to specify the analytical function for the data (default: None)
--over-samp-ratio	over sampling ratio; factor for how many points to be used in calculations (default: 2)
--verify-over-samp-ratio	over sampling ratio for verification; factor for how many points to be used in calculations (default: 0.5)
-b, --backend	the backend that will be used for Matplotlib plotting (default: ‘TkAgg’)
--conv-threshold-percent	the decimal percent of the response mean to be used as a threshold for tracking convergence (default: 0.0005)
--epist-samp-size	the number of times to sample for each variable with epistemic uncertainty (default: 125)
--aleat-samp-size	the number of times to sample for each varaible with aleatory uncertainty (default: 25000)
--epist-sub-samp-size	the number of curves to check the new high and low intervals at for a set of curves (default: 25)
--aleat-sub-samp-size	the number of samples to check the new high and low intervals at for each individual curve (default: 5000)
--sigfigs	the number of significant figures to output in the output files (default: 5)
--alpha-out	the alpha parameter used for the F-distribution threshold for values to be removed to the model (default: 0.15)

Example

When executing UQPCE as a script, options can be specified as

When executing UQPCE as a PCE object, options can be specified as

from uqpce import PCE
pce = PCE(case='casename')

[runnervmyg876:02630] mca_base_component_repository_open: unable to open mca_btl_openib: librdmacm.so.1: cannot open shared object file: No such file or directory (ignored)

Both of these examples will set the case name of the model to 'casename' and name the output directories accordingly.

Flags

Option	Description
-h, --help	show help message and exit
-v, --version	displays the version of the software
--verbose	increase output verbosity
--verify	allows verification of results
--plot	generates factor vs response plots, pbox plot, and error plots
--track-convergence-off	allows users to converge on confidence interval until the change between the two iterations is less than the threshold
--generate-samples	generates the samples used for all variables according to the parameters provided in the input file
--model-conf-int	includes uncertainties associated with the model itself
--stats	perform additional statistics for a more-comprehensive profile of the model
--report	generates a user-friendly report that documents the statistics and plots that suggest the model is insufficient
--seed	if UQPCE should use a seed for random values
--backward-elimination	if UQPCE should use backward elimination after building the model

Example

When executing UQPCE as a script, options can be specified as

When executing UQPCE as a PCE object, flags can be specified as

from uqpce import PCE
pce = PCE(version=True)

Both of these examples will output the version of UQPCE that is being used.