Re: [Numpy-discussion] making numpy.dot faster
Hi, This seems to tell that numpy has been build without altas. Hum, maybe we need to work with the Debian guys to make sure that numpy is available with altas. we had recently a discussion regarding this issue on this mailinglist, see: http://groups.google.com/group/Numpy-discussion/browse_thread/thread/507e7722f99406fa/ and on the debian bug tracker: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=489253 in debian testing and unstable numpy is now built with a patch that enables atlas support. I don't know the status of the ubuntu package, but I presume they may apply a similar patch until the numpy building system's checks are relaxed ina way that no patch is needed anymore. cheers, tiziano ___ Numpy-discussion mailing list Numpy-discussion@scipy.org http://projects.scipy.org/mailman/listinfo/numpy-discussion
Re: [Numpy-discussion] making numpy.dot faster
On Sat, Oct 04, 2008 at 05:59:40PM +0200, Tiziano Zito wrote: Hi, This seems to tell that numpy has been build without altas. Hum, maybe we need to work with the Debian guys to make sure that numpy is available with altas. we had recently a discussion regarding this issue on this mailinglist, see: http://groups.google.com/group/Numpy-discussion/browse_thread/thread/507e7722f99406fa/ and on the debian bug tracker: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=489253 OK, thanks. I thought I had seen something like this go by, but I wasn't sure. in debian testing and unstable numpy is now built with a patch that enables atlas support. I don't know the status of the ubuntu package, but I presume they may apply a similar patch until the numpy building system's checks are relaxed ina way that no patch is needed anymore. I'll can make sure that this patch makes its way into next ubuntu. Thanks a lot for recalling this discussion, and forgive me for not paying enough attention. My attention span is limited because I am doing too many things. Gaël ___ Numpy-discussion mailing list Numpy-discussion@scipy.org http://projects.scipy.org/mailman/listinfo/numpy-discussion
[Numpy-discussion] making numpy.dot faster
I am doing a calculation where one call numpy.dot ends up taking 90% of the time (the array is huge: (61373, 500) ). Any chance I can make this faster? I would believe BLAS/ATLAS would be behind this, but from my quick analysis (ldd on numpy/core/multiarray.so) it doesn't seem so. Have I done something stupid when building numpy (disclaimer: I am on a system I don't know well --Mandriva--, so I could very well have done something stupid). Cheers, Gaël ___ Numpy-discussion mailing list Numpy-discussion@scipy.org http://projects.scipy.org/mailman/listinfo/numpy-discussion
Re: [Numpy-discussion] making numpy.dot faster
On Fri, Oct 3, 2008 at 10:59 AM, Gael Varoquaux [EMAIL PROTECTED] wrote: I am doing a calculation where one call numpy.dot ends up taking 90% of the time (the array is huge: (61373, 500) ). Any chance I can make this faster? I would believe BLAS/ATLAS would be behind this, but from my quick analysis (ldd on numpy/core/multiarray.so) it doesn't seem so. Have I done something stupid when building numpy (disclaimer: I am on a system I don't know well --Mandriva--, so I could very well have done something stupid). What does np.__config__.show() show? What exactly are you multiplying? What is the original problem? Chuck ___ Numpy-discussion mailing list Numpy-discussion@scipy.org http://projects.scipy.org/mailman/listinfo/numpy-discussion
Re: [Numpy-discussion] making numpy.dot faster
Fri, 03 Oct 2008 18:59:02 +0200, Gael Varoquaux wrote: I am doing a calculation where one call numpy.dot ends up taking 90% of the time (the array is huge: (61373, 500) ). Any chance I can make this faster? I would believe BLAS/ATLAS would be behind this, but from my quick analysis (ldd on numpy/core/multiarray.so) it doesn't seem so. Have I done something stupid when building numpy (disclaimer: I am on a system I don't know well --Mandriva--, so I could very well have done something stupid). AFAIK, multiarray.so is never linked against ATLAS. The accelerated dot implementation is in _dotblas.so, and can be toggled with alterdot/ restoredot (but the ATLAS one should be active by default). numpy.dot.__module__ 'numpy.core._dotblas' Are your arrays appropriately contiguous? Numpy needs to copy the data if they are not; though I'm not sure if this could account for what you see. -- Pauli Virtanen ___ Numpy-discussion mailing list Numpy-discussion@scipy.org http://projects.scipy.org/mailman/listinfo/numpy-discussion