[Numpy-discussion] My GSoC Proposal to Implement a Subset of NumPy for PyPy
Mon Apr 19 11:49:29 CDT 2010
Dan Roberts wrote:
> Hi everybody, my name is Dan Roberts, and my Google Summer of Code
A few years back, someone did a tiny micro-prototype of numpy
implemented in Psyco, using the array.array data type as the base. It
was actually pretty promising performance-wise. I don't know if there is
anything there that would be useful, but I'd do a bit of googling and
see if you can find it.
Dag Sverre Seljebotn wrote:
> Eventually, for this to be at all useful for the NumPy crowd, one has to
> make available eigenvalue finders, FFTs, and so on as well.
I'm not sure I agree -- here I find I rarely use any of that "standard"
stuff -- not sure why, maybe I do weird work.
> In my opinion, the *primary* reason Python is used for scientific
> programming rather than some other language is how easy it is to connect
> with C, C++ and Fortran code in CPython. That's something to keep in mind.
Well, yes, this really is key -- performance aside, there are a lot of
really useful libs out there -- I don't want to have to re-write them.
Another note: I only glanced at your proposal, but it seems that one
thing you may want to start with is getting PyPy to "understand" PEP
3118 buffers, from there, you can write numpy in python, and you should
be able to use C numpy as well, passing the buffers back and forth (and
interface with arbitrary C code)
Christopher Barker, Ph.D.
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
More information about the NumPy-Discussion