Re: [Numpy-discussion] [review] Easy win to improve numpy import times by 30 %

2009-10-12 Thread David Cournapeau
On Sat, Oct 10, 2009 at 2:49 PM, David Cournapeau wrote: > On Sat, Oct 10, 2009 at 1:24 AM, Peter > wrote: >> On Fri, Oct 9, 2009 at 5:56 AM, David Cournapeau >> wrote: >>> >>> Since inspect is used in quite a few places, and that we only use it to >>> extract arguments from a function, I added

Re: [Numpy-discussion] [review] Easy win to improve numpy import times by 30 %

2009-10-09 Thread Peter
On Fri, Oct 9, 2009 at 5:56 AM, David Cournapeau wrote: > > Since inspect is used in quite a few places, and that we only use it to > extract arguments from a function, I added a small numpy.lib.inspect > module, and ... Is numpy.lib intended as a public API? How about numpy.lib._inspect instead

Re: [Numpy-discussion] [review] Easy win to improve numpy import times by 30 %

2009-10-09 Thread Chris Barker
David Cournapeau wrote: > This branch improves numpy import times quite significantly on slow > machines: > I think it largely worths it, and will integrate this unless someone is > strongly against it or see a problem with the approach, +1 -Chris -- Christopher Barker, Ph.D. Oceanographer

[Numpy-discussion] [review] Easy win to improve numpy import times by 30 %

2009-10-08 Thread David Cournapeau
Hi, This branch improves numpy import times quite significantly on slow machines: http://github.com/cournape/numpy/tree/noinspect One of the main culprit is ma, because of inspect (inspect is extremely slow to import; as a data point, python -c "import inspect" takes 67 ms vs python -c "" ta