Julien Schueller

Results 131 issues of Julien Schueller

### What is the idea? reuse MetaModelResult in Linear/Quadratic|Taylor/LeastSquares metamodels MetaModelResult should be moved to /Base maybe residuals should be computed automatically in the ctor, see also #2455 ### Why...

enhancement
API

### What is the idea? - ctor from a single Sample - computeSpearmanCorrelation->rank - computePearsonCorrelation -> linear ### Why is this needed? _No response_ ### Additional Context _No response_

enhancement
API

### What happened? we cannot see the colorbar ### How to reproduce the issue? ```shell import openturns as ot field = ot.GaussianProcess(ot.ExponentialModel([1.0]*2, [1.0]), ot.IntervalMesher([20]*2).build(ot.Interval([-10.0]*2,[10.0]*2))).getRealization() ot.Show(field.drawMarginal()) ``` ### Version 1.21 ###...

bug
graph

### What is the idea? is there a goodness of fit test for copulas available we could implement in OT ? currently we only have BIC that's a continuous measure...

enhancement

### What is the idea? When performing a BIC test with a copula, one must normalize the sample into [0,1] else BIC returns MaxScalar. I think that would deserve some...

enhancement

the cmake version could be bumped to 3.5 to avoid this deprecation message ``` CMake Deprecation Warning at CMakeLists.txt:1 (cmake_minimum_required): Compatibility with CMake < 3.5 will be removed from a...

enhancement

FMPy does not build with sundials 6.x, here is a small patch to add the necessary context argument when calling sundials functions: [0001-Compile-with-sundials-6.patch.txt](https://github.com/CATIA-Systems/FMPy/files/13256526/0001-Compile-with-sundials-6.patch.txt) this is necessary to get the full...

the printf/fprintf format argument is missing in several places: ``` diff --git a/src/modelica/ModelicaFMI.c b/src/modelica/ModelicaFMI.c index a74cd73..a1fca6d 100644 --- a/src/modelica/ModelicaFMI.c +++ b/src/modelica/ModelicaFMI.c @@ -48,7 +48,7 @@ static void logFunctionCall(FMIInstance* instance, FMIStatus...

the gui part uses PyQt, and the license text does not reflect the GPL our users were tricked into believing they can redistribute derived works of fmpy into BSD seems...

I also stumbled on #129, one possibility to address this is to raise at the import Maybe we could also just send a warning message, maybe if there are legitimate...