Status: Accepted
Owner: [email protected]
Labels: Type-Enhancement Priority-Low
New issue 3061 by [email protected]: Stats sampling using numpy
http://code.google.com/p/sympy/issues/detail?id=3061
The stats module has a sampling mechanism to provide approximate solutions
to statistical queries. This mechanism is relatively slow and 1000+ sample
simulations of complex expressions can take several seconds.
X = Normal(0,1)
Y = Exponential(10)
Var(sin(X+Y)) # this hangs because the integral is tough
<Ctrl-C>
Var(sin(X+Y), numsamples=1000) # this takes a few seconds
0.420095049234496
NumPy also provides a sampling mechanism in its random module. If we could
sample large arrays out of the underlying distributions and then push those
large arrays through the sympy expressions we could compute approximate
solutions much more quickly.
import numpy
xs = numpy.random.normal(0,1, 1000)
ys = numpy.random.exponential(10, 1000)
numpy.sin(xs+ys).var() # This takes 150 micro seconds
0.48987452637810897
So, can we use sympy to build up a complex statistical question and then
use numpy to compute the result?
--
You received this message because you are subscribed to the Google Groups
"sympy-issues" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sympy-issues?hl=en.
