[Scipy-tickets] [SciPy] #1765: scipy.special.hyp2f0(1.0, 0.5, -10, c)

SciPy Trac scipy-tickets@scipy....
Sat Nov 3 15:37:29 CDT 2012


#1765: scipy.special.hyp2f0(1.0, 0.5, -10, c)
------------------------------+---------------------------------------------
 Reporter:  unknown (google)  |       Owner:  pv         
     Type:  defect            |      Status:  new        
 Priority:  normal            |   Milestone:  Unscheduled
Component:  scipy.special     |     Version:  0.10.1     
 Keywords:  hyp2f0            |  
------------------------------+---------------------------------------------
 This scipy.special function takes an extra convergence factor type c and
 returns an evaluation and an error estimate.  For both convergence factor
 types c=1 and c=2, scipy.special.hyp2f0(1.0, 0.5, -10, c) gives
 unnecessarily uninformative results.

 >>> scipy.special.hyp2f0(1.0, 0.5, -10, 1)
 (-27.75, 75.0)

 >>> scipy.special.hyp2f0(1.0, 0.5, -10, 2)
 (-4.3333333333333321, 75.0)


 For negative x, this function could use a transformation to
 scipy.special.hyperu as hyp2f0(a, b, x) = (-1/x)**a * hyperu(a, 1+a-b,
 -1/x) as seems to be done by mpmath and the gnu scientific library.

 >>> mpmath.hyp2f0(1.0, 0.5, -10)
 mpf('0.40556507920419638')

 >>> 0.1*scipy.special.hyperu(1.0, 1.5, 0.1)
 0.40556507920419621

 These are the results that I would have expected from
 scipy.special.hyp2f0(1.0, 0.5, -10).

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1765>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list