[ot-users] Matern Covariance Model for Kriging Surrogate Model

LELIEVRE-Nicolas nicolas.lelievre at sigma-clermont.fr
Fri Aug 25 08:23:45 CEST 2017


Hi Sofiane and Julien, 

Thank you for your time. 
Unfortunately, I work on Windows. 
I thus will test it when it will be available on Windows. 
I am not in a hurry. 

Thanks again, 
Nicolas 


----- Mail original -----

De: "Julien Schueller | Phimeca" <schueller at phimeca.com> 
À: users at openturns.org 
Envoyé: Jeudi 24 Août 2017 22:11:02 
Objet: Re: [ot-users] Matern Covariance Model for Kriging Surrogate Model 



Hi Nicolas, 




Sofiane identified the bug. 

If you're using openturns via conda on linux/osx we could deploy a fix for this if you're interested. 




j 



De : users-bounces at openturns.org <users-bounces at openturns.org> de la part de HADDAD Sofiane <sofiane_haddad at yahoo.fr> 
Envoyé : mercredi 23 août 2017 17:57 
À : LELIEVRE-Nicolas; users at openturns.org 
Objet : Re: [ot-users] Matern Covariance Model for Kriging Surrogate Model 
Hi, 

There is indeed a bug within the MaternModel::setParameter, thanks for the report. 
It is fixed in Fix http://trac.openturns.org/ticket/905 by sofianehaddad · Pull Request #537 · openturns/openturns 

Sofiane 
	


	Fix http://trac.openturns.org/ticket/905 by sofianehaddad · Pull Request #5... 
MaternModel::setFullParameter should update all internal parameters and check the size/accuracy of the input arg... 
	








Le Mardi 22 août 2017 13h51, HADDAD Sofiane <sofiane_haddad at yahoo.fr> a écrit : 


Hi Nicolas, 

Sorry for the delay. 

I will have a look at the problem this afternoon 


Regards, 
Sofiane 


Le Jeudi 17 août 2017 15h33, LELIEVRE-Nicolas <nicolas.lelievre at sigma-clermont.fr> a écrit : 


Hi, 

I want to calibrate a Kriging surrogate model in OpenTurns and I face difficulties. 
Indeed, I want to use the Matern covariance model. 
But, when I run the KrigingAlgorithm optimization, the scale parameters (theta) are not optimized. 
I have studied the problem and found that the LogLikelihood function is constant, no matter how points are in the DoE, what the performance function is and what the dimension is. 

I think that the problem is on the definition of the covariance model since if I use SquaredExponential there is not any problems. 
But, I don't find how to define it correctly. 

May you provide me some helpful advice ? 
Thank you in advance. 

A little example: 

import numpy as np 
import openturns as ot 
def G(X): 
out = 15 - (X[:,0]**2 + X[:,1]**2 - 5*np.cos(2*np.pi*X[:,0]) - 5*np.cos(2*np.pi*X[:,1])) 
return out 

dim = 2 
Loi = np.ones(dim) 
Moy = np.ones(dim) 
Stdev = np.ones(dim) 
nini = 100 

nva = np.size(Loi) 
DOE_u = np.random.normal(0,1,(nini,nva)) 
DOE_y = G(DOE_u) 
DOE_y = DOE_y.reshape((nini,1)) 

inputSample = ot.Sample(DOE_u) 
outputSample = ot.Sample(DOE_y) 
basis = ot.ConstantBasisFactory(nva).build() 
covarianceModel = ot.MaternModel(nva) 
covarianceModel.setNu(5/2) 
algo = ot.KrigingAlgorithm(inputSample, outputSample, covarianceModel, basis) 
algo.run() 
result = algo.getResult() 
print(result.getCovarianceModel()) 
LogLikelihood = algo.getReducedLogLikelihoodFunction() 

Nicolas Lelièvre 
Doctorant Institut Pascal Clermont-Ferrand 

_______________________________________________ 
OpenTURNS users mailing list 
users at openturns.org 
http://openturns.org/mailman/listinfo/users 





_______________________________________________ 
OpenTURNS users mailing list 
users at openturns.org 
http://openturns.org/mailman/listinfo/users 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://openturns.org/pipermail/users/attachments/20170825/4054e4e7/attachment.html>


More information about the users mailing list