otaf.sensitivity package

Module contents

class otaf.sensitivity.SobolIndicesExperimentWithComposedDistribution(composedDistribution=None, size=None, second_order=False)

Bases: SobolIndicesExperiment

generate(**kwargs)

Generates and returns the final mixture matrix.

Keyword Arguments:

• method (str) – Can be : ‘MonteCarlo’, ‘LHS’, ‘QMC’

• sequence (str) – Only if using QMC Can be : ‘Faure’, ‘Halton’, ‘ReverseHalton’, ‘Haselgrove’, ‘Sobol’

generateWithWeights(**kwargs)

Not implemented, for coherence with openturns library

getClassName()

Returns the name of the class.

getId()

Returns the ID of the object.

getName()

Returns the name of the object.

getShadowedId()

Returns the shadowed ID of the object.

getSize()

Returns the size of the generated mixture matrix.

getVisibility()

Returns the visibility

hasName()

Returns if the object has a name

hasUniformWeights()

Not implemented, for coherence with openturns library

hasVisibleName()

Returns if yes or not the name is visible

setComposedDistribution(composedDistribution)

Sets the composed distribution.

setName(name)

Sets the name of the object

setShadowedId(ids)

Sets the shadowed ID of the object.

setSize(N)

Sets the size of the samples A and B.

otaf.sensitivity.plotSobolIndicesWithErr(S, errS, varNames, n_dims, Stot=None, errStot=None, dimNames=None, figsize=(20, 10))

Function to plot the Sobol’ indices with an errorbar to visualize the uncertaintity in the estimator. (only first and total order yet)

Note

Function is written to adapt the plotting according to the dimensions of the input, for higher level inputs, mayavi seems the most reasonable choice for plotting