On Oct 4, 2009, at 10:07 AM, mark mcclure wrote: > I'm investigating the complex dynamics of Airy functions with > Python and I wonder if I can speed up the process significantly > with Cython. I've successfully sped up the code by about 50% > but I might expect much greater speed improvement. I'm > guessing the problem is that Cython can't deal with SciPy's > complex valued airy function. > 1) Am I likely correct that the airy function is the > culprit?
It's certainly possible, but see the comments below. > 2) If so, is it likely that Cython will be eventually > expanded to include the complex valued special functions > of SciPy? No, but you might want to call these functions in SciPy directly as C functions. (You'd have to look up the SciPy headers to see what to call them.) > If you're interested in the specific code, you can view it > as a Sage notebook here: > http://sagenb.org/home/pub/1013/ http://sagenb.org/home/pub/1013/cells/1/ __home_sage_sagenb_sage_notebook_worksheets_mcmcclur_8_code_sage7_spyx.h tml Lines 27, 31, 33: You're calling abs as a Python function. Use cabs, which can be imported as cdef extern from "complex.h": double cabs(double complex) Line 24-25 has a huge amount of Cython code in it. complex(...) and float(...) create *Python* complex and float objects. Click on those lines to see the insane amount of Python going on. Try writing z =(((xmax - xmin)*j)/resRe + xmin) + ((ymax - ymin)*k)/resIm + ymin) * 1.0j instead. You should be getting much more than a 50% speedup. - Robert --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-support@googlegroups.com To unsubscribe from this group, send email to sage-support-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-support URL: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
