Thanks! The problem came from the 1/2 ! For convenience I've found the
function digamma on numpy

*http://docs.scipy.org/doc/scipy/reference/generated/scipy.special.psi.html

But it's quite hard to find it! Maybe we can ask to add digamma in the
title between parenthesis?


Fabien
*
2012/7/10 Damon McDougall <damon.mcdoug...@gmail.com>

> On Tue, Jul 10, 2012 at 08:57:24AM -0400, Benjamin Root wrote:
> > On Tue, Jul 10, 2012 at 7:05 AM, Damon McDougall
> > <damon.mcdoug...@gmail.com>wrote:
> >
> > > On Tue, Jul 10, 2012 at 12:27:59PM +0200, Fabien Lafont wrote:
> > >
> > > > But It returns zero division error even when x is in ]0,1]
> > >
> > > I think it blows up at x = 0. What is the type of x in your usecase? Is
> > > it an array? If x contains the element 0, you will get a zero
> > > division error. You could try plotting the points explicitly:
> > >
> > Another problem might be the "1/2" part, which in python2.x would yield 0
> > unless one does "from __future__ import division".
> >
> > Ben Root
>
> Wow, I can't believe I didn't spot that. Nice one.
>
> I will update my answer according to Ben's astute observation:
>
> from scipy import special
> from pylab import *
>
> x = linspace(0.5, 2.0, num=100, endpoint=True)
> y = special.polygamma(0, 0.5 + 1.0/x)
> plot(x, y)
> show()
>
> Thanks Ben.
>
> --
> Damon McDougall
> http://damon-is-a-geek.com
> B2.39
> Mathematics Institute
> University of Warwick
> Coventry
> West Midlands
> CV4 7AL
> United Kingdom
>
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Reply via email to