Comments on: Easy Python/Numpy CUDA/CUBLAS Integration

By: Boom

Boom — Fri, 03 Jun 2011 15:34:29 +0000

Hi can you post the code for CPU and GPU I would like to test it on my Supermicro server with a Tesla card in.

By: sean

sean — Wed, 16 Mar 2011 01:07:19 +0000

I ran into the same bug as Bob (change 190-191 from “if” to “elif”).

By: derek

derek — Fri, 05 Mar 2010 03:24:57 +0000

sorry, link fixed!

By: arief nur andono

arief nur andono — Fri, 26 Feb 2010 04:36:20 +0000

why pycublaspy1.txt no longer exist in this website? where should i download it??

By: Ashleigh

Ashleigh — Tue, 09 Jun 2009 06:11:32 +0000

First of all, thanks for sharing this. That said I am having some trouble because I have a 64bit Mac and I think* that my CUDA install expects 64bit, do you know how hard it would be to make your current code 64 bit compatible? I’m planning to tinker with it myself and see if I can get it working but if you could offer any advice I would really appreciate it.
Cheers!

*
>>> A = numpy.mat([[1,2,3],[4,5,6]], numpy.float32)
>>> B = numpy.mat([[2,3],[4,5],[6,7]], numpy.float32)
>>> C = (CUBLASMatrix(A)*CUBLASMatrix(B)).np_mat()
>>> C
Warning: invalid value encountered in reduce
Warning: invalid value encountered in reduce
array([[ nan, nan],
[ nan, 79. ]], dtype=float32)

By: Bob

Bob — Fri, 17 Apr 2009 01:05:10 +0000

Thanks, appreciate this! Think you need to change 190-191 from “if” to “elif”, for non-linux machines.

By: derek

derek — Tue, 14 Apr 2009 16:40:36 +0000

i don't know much about FFTs, but it would seem that CUFFT doesn't provide that much of a performance increase. at least according to http://www.science.uwaterloo.ca/~hmerz/CUDA_benchFFT/. but if you do want to try and use it, the numpy -> graphics card code in the __init__() function should help you in using the APIs. just need to load the CUFFT library and create ctypes bindings to it.

By: Ahmed Fasih

Ahmed Fasih — Mon, 13 Apr 2009 21:24:20 +0000

This is great! Any advice on how appropriate reusing your efforts for CUFFT would be?