Régis LEBRUN

Results 22 issues of Régis LEBRUN

I can see two issues in the way the Nataf transformation is defined In Transformation.py: * The Nataf transformation is presented as a way to induce correlation between random variables,...

See in [this preprint](https://deepai.org/publication/numerical-issues-in-maximum-likelihood-parameter-estimation-for-gaussian-process-regression) an interesting comparison of different Python kriging packages on the Branin test case. It is OT 1.13 which is tested, and the results are not so...

enhancement

In ComposedDistribution::get(Inverse)IsoProbabilisticTransformation the parameters of the marginal transformation include the parameters of the copula, and the copula transformation does not include any parameters. #554 attempted to provide MarginalTransformationEvaluation::getParameters but the...

needs review

``` UnsignedInteger NumericalMathFunction::getCallsNumber() const { return getEvaluationCallsNumber(); } ``` Is it necessary to keep both? Linked to [http://trac.openturns.org/ticket/733](http://trac.openturns.org/ticket/733)

needs review

I made a little test : * g = r-s with 2 uniform distributions without any common point r U(2,3) and s U(0,1) * I use the Cobyla algorithm I...

needs review

rationale: the graphs must be part of the result, as in optimization algo/result * copy convergence history * deprecate draw methods Linked to [http://trac.openturns.org/ticket/956](http://trac.openturns.org/ticket/956)

API

For now the `HistoryStrategy` stores the sample as an attribute, hence in RAM which is dumped in the void in case of an error, either in the lib or in...

enhancement

Improved TruncatedNormal. Now it is possible to truncate a normal distribution on more extreme truncation intervals. It fixes #2058.

enhancement

When m==a or m==b, there is a division by 0 in this method: ``` import openturns as ot d1 = ot.Triangular(-2.0, -2.0, 4.0) print(d1.computeCharacteristicFunction(1.0)) d2 = ot.Triangular(-2.0, 4.0, 4.0) print(d2.computeCharacteristicFunction(1.0))...

bug

In [the documentation of KarhunenLoeveSVDAlgorithm](https://openturns.github.io/openturns/latest/user_manual/_generated/openturns.KarhunenLoeveSVDAlgorithm.html) it is said that this algorithm is a special case of KarhunenLoeveQuadratureAlgorithm, which is wrong: it is a special case of KarhunenLoeveP1Algorithm. The documentation refers...

doc