1%feature("docstring") OT::MetaModelAlgorithm 2"Base class to compute a metamodel. 3 4Available constructor: 5 MetaModelAlgorithm(*distribution, model*) 6 7Parameters 8---------- 9distribution : :class:`~openturns.Distribution` 10 Joint probability density function of the physical input vector. 11model : :class:`~openturns.Function` 12 Physical model to be approximated by a metamodel. 13 14Notes 15----- 16A MetaModelAlgorithm object can be used only through its derived classes: 17 18- :class:`~openturns.KrigingAlgorithm` 19 20- :class:`~openturns.FunctionalChaosAlgorithm`" 21 22// --------------------------------------------------------------------- 23 24 25%feature("docstring") OT::MetaModelAlgorithm::BuildDistribution 26"Recover the distribution, with metamodel performance in mind. 27 28For each marginal, find the best 1-d continuous parametric model 29else fallback to the use of a nonparametric one. 30 31The selection is done as follow: 32 33 - We start with a list of all parametric models (all factories) 34 - For each model, we estimate its parameters if feasible. 35 - We check then if model is `valid`, ie if its Kolmogorov score exceeds a threshold 36 fixed in the `MetaModelAlgorithm-PValueThreshold` ResourceMap key. Default value is 5% 37 - We sort all `valid` models and return the one with the optimal criterion. 38 39For the last step, the criterion might be `BIC`, `AIC` or `AICC`. The specification of the criterion is 40done through the `MetaModelAlgorithm-ModelSelectionCriterion` ResourceMap key. Default value is fixed to `BIC`. 41Note that if there is no `valid` candidate, we estimate a non-parametric model (:class:`~openturns.KernelSmoothing` 42or :class:`~openturns.Histogram`). The `MetaModelAlgorithm-NonParametricModel` ResourceMap key allows selecting 43the preferred one. Default value is `Histogram` 44 45One each marginal is estimated, we use the Spearman independence test on each component pair to decide whether an 46independent copula. In case of non independence, we rely on a :class:`~openturns.NormalCopula`. 47 48Parameters 49---------- 50sample : :class:`~openturns.Sample` 51 Input sample. 52 53Returns 54------- 55distribution : :class:`~openturns.Distribution` 56 Input distribution." 57// --------------------------------------------------------------------- 58 59%feature("docstring") OT::MetaModelAlgorithm::getDistribution 60"Accessor to the joint probability density function of the physical input vector. 61 62Returns 63------- 64distribution : :class:`~openturns.Distribution` 65 Joint probability density function of the physical input vector." 66 67// --------------------------------------------------------------------- 68 69%feature("docstring") OT::MetaModelAlgorithm::setDistribution 70"Accessor to the joint probability density function of the physical input vector. 71 72Parameters 73---------- 74distribution : :class:`~openturns.Distribution` 75 Joint probability density function of the physical input vector." 76 77// --------------------------------------------------------------------- 78 79%feature("docstring") OT::MetaModelAlgorithm::getInputSample 80"Accessor to the input sample. 81 82Returns 83------- 84inputSample : :class:`~openturns.Sample` 85 Input sample of a model evaluated apart." 86 87// --------------------------------------------------------------------- 88 89%feature("docstring") OT::MetaModelAlgorithm::getOutputSample 90"Accessor to the output sample. 91 92Returns 93------- 94outputSample : :class:`~openturns.Sample` 95 Output sample of a model evaluated apart." 96 97// --------------------------------------------------------------------- 98 99%feature("docstring") OT::MetaModelAlgorithm::run 100"Compute the response surfaces. 101 102Notes 103----- 104It computes the response surfaces and creates a 105:class:`~openturns.MetaModelResult` structure containing all the results."