[Scipy-tickets] [SciPy] #1510: Add methods to compute derivatives to Scipy

SciPy Trac scipy-tickets@scipy....
Sun Oct 30 11:52:35 CDT 2011


#1510: Add methods to compute derivatives to Scipy
------------------------------------------------------------------------------------------------------+
 Reporter:  deil                                                                                      |       Owner:  somebody   
     Type:  enhancement                                                                               |      Status:  new        
 Priority:  normal                                                                                    |   Milestone:  Unscheduled
Component:  Other                                                                                     |     Version:  0.9.0      
 Keywords:  derivative, numerical derivative, algorithmic derivative, Jacobian, Hessian,  scipy.diff  |  
------------------------------------------------------------------------------------------------------+

Comment(by rgommers):

 I agree that it would be useful to have better tools in scipy to compute
 derivatives.  This probably deserves its own discussion on scipy-dev, the
 previous mention got lost in the scipy.optimize interface thread.

 Looking at the code, it seems to me that algopy is the package that is
 more polished and also probably more robust/performant. On the other hand,
 it's a little harder to use. So there may be a place for both in scipy, if
 the authors of both packages agree with that. The packages may benefit
 from more users. Numdifftools would require some work I think before it
 could be included.

 I'm not so worried about code size, both packages are small and pure
 Python. Code quality, documentation and test coverage matter more than
 size.

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1510#comment:4>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list