On 22/11/2016 12:34, Skip Montanaro wrote:
I'm simply suggesting there is plenty of room for improvement. I even
showed a version that did *exactly* what numpy does (AFAIK) that was three
times the speed of numpy even executed by CPython. So there is some mystery
there.
As I indicated in my earlier response, your version doesn't pass all of
numpy's cross product unit tests. Fix that and submit a patch to the numpy
maintainers. I suspect it would be accepted.
I saw your response but didn't understand it. My code was based around
what Peter Otten posted from numpy sources.
I will have a look. Don't forget however that all someone is trying to
do is to multiply two vectors. They're not interested in axes
transformation or making them broadcastable, whatever that means.
So making numpy.cross do all that may simply be too big a cost.
--
bartc
--
https://mail.python.org/mailman/listinfo/python-list
