Dear All,
I'm trying to integrate a multidimensional Gaussian with sympy. However,
when I perform the integration, it crashes with an error related to
polar_lift:
import sympy as sy
from sympy.abc import *
sy.init_printing()
C= sy.MatrixSymbol("C", 3, 3)
r = sy.IndexedBase("r", real = True)
a,b,c = sy.symbols("a,b,c", integer = True)
integ = sy.Integral( sy.exp(-r[a] * C[a,a] * r[a] - 2*r[a] * C[a,b] * r[b]
- r[b]*C[b,b] * r[b]),
(r[a], -sy.oo, sy.oo), (r[b], -sy.oo, sy.oo))
integ.doit()
In the last line, the code crashes with the following error
~/anaconda2/lib/python3.7/site-packages/sympy-1.6.dev0-py3.7.egg/sympy/matrices/expressions/matexpr.py
in name(self) 797 @property 798 def name(self):--> 799
return self.args[0].name 800 801 def _eval_subs(self, old, new):
AttributeError: 'polar_lift' object has no attribute 'name'
If I replace the indexed variables with normal variables, the integration
is performed (even if it is superslow, and if I ask for conds="none" it
assumes it to diverge)
There is a way to perform integrals of multidimensional Gaussians? In
principle, these kinds of integrals should be tabulated
Thanks for your attention,
Bests,
Lorenzo
--
You received this message because you are subscribed to the Google Groups
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/1ea82042-cd83-4dc0-86fe-c64d8cc1006d%40googlegroups.com.
