dubourg/python-randomfields

A Python module that implements tools for the simulation and identification of random fields using the Karhunen-Loeve expansion representation.

python-randomfields's Issues

Question for 2D asymmetric space

Hello.
I want to know how the code could be modified if the random space is asymmetric for KL expansion.
Thanks a lot!

TypeError

Hello, when I was running this code ‘1D_karhunen_loeve_simulation_example.py’, it reported the following error. I don't understand it. What do I need to do?

D:\anaconda3\lib\site-packages\openturns\func.py in init(self, *args)
1112
1113 def init(self, *args):
-> 1114 _func.EnumerateFunction_swiginit(self, _func.new_EnumerateFunction(*args))
1115 swig_destroy = _func.delete_EnumerateFunction
1116

TypeError: Wrong number or type of arguments for overloaded function 'new_EnumerateFunction'.
Possible C/C++ prototypes are:
OT::EnumerateFunction::EnumerateFunction()
OT::EnumerateFunction::EnumerateFunction(OT::EnumerateFunctionImplementation const &)
OT::EnumerateFunction::EnumerateFunction(OT::EnumerateFunction::Implementation &)
OT::EnumerateFunction::EnumerateFunction(OT::EnumerateFunction const &)

TypeError: OrthogonalProductPolynomialFactory_getNodesAndWeights() takes exactly 2 arguments (3 given)

As of openturns 1.4, reference interfaces have been removed, so that there is no more need for passing weights to the getNodesAndWeights method.

Bugs for polynomial nodes

Hello,

I have tried to used the code for my database and it seems to me that there is an error relating to nodes of polynomials.

    U, W = polynoms.getNodesAndWeights(
        ot.Indices([legendre_quadrature_order] * dimension * 2))

would create nodes U independent of the domain of definition. Thus

    X = np.repeat(scale, 2) * U + np.repeat(shift, 2)

would create X which exceed the domain of definition. In consequence, we would get nan for matrix C.

Could you please try to fix this problem? Thank you in advance.

missing sample files

2D_sample_paths.txt
1D_sample_paths.txt