Régis LEBRUN

Results 22 issues of Régis LEBRUN

Many of the methods of the classes related to OptimalLHSExperiment have no const qualifiers for their arguments, which breaks the coding rules. Here is a first list: ``` Scalar TemperatureProfile::operator()(UnsignedInteger...

enhancement
API

Hi, In order to ease the use of ResourceMap, I propose to automatically convert the String entries into upper-case (done in ResourceMap::SetAsString/ResourceMap::addAsString) and to modify the C++ code accordingly. It...

enhancement

The formula used to compute the P1 gram matrix of a mesh has been fixed in Mesh, but not in KarhunenLoeveP1Algorithm::ComputeSparseAugmentedP1Gram. BTW, many other small or big issues have to...

needs review
Performance

The reference to ResourceMap keys is not uniform within the documentation: ``` python/src/ExpectationSimulationAlgorithm_doc.i.in:The default type is set by the `ExpectationSimulationAlgorithm-DefaultCoefficientOfVariationCriterionType` key of the :class:`~openturns.ResourceMap`. python/src/MultiStart_doc.i.in: Default value is *MultiStart-KeepResults* in...

enhancement
doc

### What happened? 1. See https://openturns.github.io/openturns/latest/theory/probabilistic_modeling/copulas.html ### How to reproduce the issue? ```shell See https://openturns.github.io/openturns/latest/theory/probabilistic_modeling/copulas.html ``` ### Version 1.23dev ### Operating System Linux ### Installation media from source ### Additional...

bug

### What happened? Trying to install openturns using pip results into an error with python 3.12 ### How to reproduce the issue? ```shell pip install openturns ``` ### Version 1.22...

help wanted

Hi, Many thanks for this list! Are you aware of the hmat-oss library? It provides several formats and compression algorithms, and IMO it deserves to should be added to the...

Improved KernelSmoothing Now, a log transformation can be applied to ease the reconstruction of skewed distributions. This new capability has been thoroughly documented thanks to @adutfoy

The following distributions are renamed: * ConditionalDistribution becomes DeconditionedDistribution * BayesDistribution becomes JointByConditioningDistribution * NegativeBinomial becomes Polya * Spurious uses of ComposedDistribution have been fixed

I moved the mechanism already proposed in LeastSquaresDistributionFactory, MaximumLikelihoodFactory, QuantileMatchingFactory, MethodOfMomentsFactory at the DistributionFactoryImplementation level in order to allow the user to fix a part of the parameter values before...

enhancement
in progress
probas