On Thu, Jul 24, 2014 at 12:59 PM, Charles R Harris < charlesr.har...@gmail.com> wrote:
> > > > On Thu, Jul 24, 2014 at 8:27 AM, Jaime Fernández del Río < > jaime.f...@gmail.com> wrote: > >> On Thu, Jul 24, 2014 at 4:56 AM, Julian Taylor < >> jtaylor.deb...@googlemail.com> wrote: >> >>> In practice one of the better methods is pairwise summation that is >>> pretty much as fast as a naive summation but has an accuracy of >>> O(logN) ulp. >>> This is the method numpy 1.9 will use this method by default (+ its >>> even a bit faster than our old implementation of the naive sum): >>> https://github.com/numpy/numpy/pull/3685 >>> >>> but it has some limitations, it is limited to blocks fo the buffer >>> size (8192 elements by default) and does not work along the slow axes >>> due to limitations in the numpy iterator. >>> >> >> For what it's worth, I see the issue on a 64-bit Windows numpy 1.8, but >> cannot on a 32-bit Windows numpy master: >> >> >>> np.__version__ >> '1.8.0' >> >>> np.ones(100000000, dtype=np.float32).mean() >> 0.16777216 >> >> >>> np.__version__ >> '1.10.0.dev-Unknown' >> >>> np.ones(100000000, dtype=np.float32).mean() >> 1.0 >> >> > Interesting. Might be compiler related as there are many choices for > floating point instructions/registers in i386. The i386 version may > effectively be working in double precision. > Also note the different numpy version. Julian told that numpy 1.9 will use a more precise version in that case. That could explain that. Fred
_______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
