[ot-users] Bug in distribution algebra

regis lebrun regis_anne.lebrun_dutfoy at yahoo.fr
Thu Oct 5 18:26:47 CEST 2017


Hi Phil,

I am very surprised by your bug: I checked your example against master (the upcoming OT 1.10), OT 1.9, OT 1.8 and OT 1.7 and it works perfectly. Could you provide me more information about your environment? I have only a Linux 64 bits box, so I may miss something...

Regards

Régis



________________________________
De : Phil Fernandes <phil.fernandes at enbridge.com>
À : Julien Schueller | Phimeca <schueller at phimeca.com>; "users at openturns.org" <users at openturns.org> 
Envoyé le : Jeudi 5 octobre 2017 18h03
Objet : Re: [ot-users] Bug in distribution algebra



I am running version 1.9, but I have disabled simplifications as you suggested and it is working correctly now. Thank you for your quick replies!
 
From:users-bounces at openturns.org [mailto:users-bounces at openturns.org] On Behalf Of Julien Schueller | Phimeca
Sent: Thursday, October 05, 2017 1:42 AM
To: users at openturns.org
Subject: [External] Re: [ot-users] Bug in distribution algebra
 
After some more checking the CDF & quantile are ok with both the 1.9 version and the current master,
what version are you running ?
 
 
Here is the script I used:
 
import openturns as ot
from openturns.viewer import View
x_dist = ot.Weibull(0.11838, 1.6767, -0.034709)
y_dist = ot.Normal(0.354331, 0.354331*0.0235)
ls_dist = x_dist - y_dist
ls_dist.drawCDF()
View(ls_dist.drawCDF()).show()
for i in range(10):
    print(ls_dist.computeCDF(-4.0/10.0+i/50.0))
  
0.00042869973445390443
0.019440693720066658
0.09355158417553083
0.2053317087432412
0.3329714740337595
0.46134585220658253
0.5801972194337283
0.6835213642762484
0.7688041869759314
0.8361003299792358
 
for i in range(10):
    print(ls_dist.computeQuantile(i/10.0))
 
[-0.452745]
[-0.358691]
[-0.340872]
[-0.325076]
[-0.309661]
[-0.293729]
[-0.27641]
[-0.256451]
[-0.231363]
[-0.1938]
 
 
j
 

________________________________

De :users-bounces at openturns.org <users-bounces at openturns.org> de la part de Julien Schueller | Phimeca <schueller at phimeca.com>
Envoyé : jeudi 5 octobre 2017 08:49:27
À : users at openturns.org
Objet : Re: [ot-users] Bug in distribution algebra 
 
Hello Phil,
 
This seems to be a bug of the RandomMixture simplifications, a workaround is to disable simplifications:
ot.ResourceMap.SetAsBool("RandomMixture-SimplifyAtoms", False)
 
Thank you for reporting, I hope for this to get fixed for the 1.10 version.
 
j

________________________________

De :users-bounces at openturns.org <users-bounces at openturns.org> de la part de Phil Fernandes <phil.fernandes at enbridge.com>
Envoyé : jeudi 5 octobre 2017 00:31:26
À : users at openturns.org
Objet : [ot-users] Bug in distribution algebra
 
I’ve come across some odd behavior when subtracting a normal distribution from a Weibull, e.g., 
x_dist = ot.Weibull(0.11838, 1.6767, -0.034709)
y_dist = ot.Normal(0.354331, 0.354331*0.0235)
ls_dist = x_dist – y_dist
 
The PDF function is correct, but the CDF and Quantile functions are nonsensical. I have uploaded figures showing this behavior here: https://imgur.com/a/GoVw3
 
Are there any caveats to be aware of when doing distribution algebra in OpenTurns? If I reverse the subtraction, i.e., y_dist – x_dist, then subtract the resulting CDF from 1, I get what appears to be the CDF I am looking for.
 
Thank you!
 
Phil Fernandes
 
_______________________________________________
OpenTURNS users mailing list
users at openturns.org
http://openturns.org/mailman/listinfo/users



More information about the users mailing list