[SciPy-Dev] Splines

Charles R Harris charlesr.harris@gmail....
Fri Feb 22 21:49:01 CST 2013


On Thu, Feb 21, 2013 at 7:46 AM, Charles R Harris <charlesr.harris@gmail.com
> wrote:

>
>
> On Thu, Feb 21, 2013 at 7:19 AM, Pauli Virtanen <pav@iki.fi> wrote:
>
>> Charles R Harris <charlesr.harris <at> gmail.com> writes:
>> [clip]
>> > > https://github.com/pv/scipy-work/commits/spline-unify
>> >
>> > That class is already pretty high level ;)
>>
>> Well, I was thinking forward on how to force spleval/UnivariateSpline
>> be equivalent, by making (t, c, k) and the spline class interchangeable.
>> I also started trying to force things into FITPACK, but soon
>> became convinced that this is not the way to go...
>>
>> I think you are aiming at a similar design as in numpy.polynomial,
>> which sounds good to me.
>>
>
> Well, it's what I'm familiar with ;) I think your work on piecewise
> polynomials fits in here somewhere also, but I haven't taken a close look
> at that yet.
>
>
I've put up a working repository at https://github.com/charris/bsplines.
There are some preliminary python routines for evaluating splines and such,
sort of pseudo C'ish which makes them really slow on account of all the
indexing, but maybe easier to cythonize. Anyone who is interested can take
a look and join in. Flaming welcome :) Also constructive comments/work.
Everything is 1-D at the moment until things settle down a bit.

Chuck
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/scipy-dev/attachments/20130222/1e8c0e9e/attachment.html 


More information about the SciPy-Dev mailing list