[SciPy-User] Speed up sin/sqrt functions with cython

[Gregor Thalhammer]

Am 12.1.2011 um 13:05 schrieb <g.plantageneto at runbox.com> <g.plantageneto at runbox.com>:

> 
> Hi,
> 
> I am using cython to speed up some computations (btw, thanks a lot to the people who gave me suggestions on a previous thread).
> I don't understand how to apply the fast C-coded sin/sqrt functions to a numpy array.
> I can import from math.h like this:
> 
> cdef extern from "math.h"
>    double sin(double)
> 
> but then I can't use this function on a numpy ndarray, obviously.
> 
> Any ideas? It sounds silly to me to write a cycle on ndarray elements.
> 
> Thanks

For sin/sqrt computations on large arrays, numexpr + VML might give you better results than a cython implementation, and with less effort. numexpr with VML uses Intels VML library which gives much better performance than a C loop, at least for sin, for sqrt the performance is similar. In addition, numexpr-VML makes use of all your CPU cores.

Gregor