Re: Modified Akima interpolation

Akima discusses a modification in the 1991 paper cited in the Matlab
documentation

https://dl.acm.org/doi/pdf/10.1145/114697.116810

At a glance, I can’t tell if it is exactly the same modification, though it
does have the property of connecting collinear points with a straight line.
I don’t have time to check carefully now.

On Tue, Apr 11, 2023 at 09:12 Pamphile Roy <[email protected]> wrote:

> Hi everyone,
>
> Thank you for the proposal. It sounds interesting and yes a keyword in the
> init sounds better in this case.
>
> But could you point out to a reference in the literature? From a very
> quick search I did not find much about this modified version. One criteria
> for inclusion in SciPy is a proven track record in the literature.
>
> Cheers,
> Pamphile
>
>
> On 11.04.2023, at 14:58, Joseph Fox-Rabinovitz <[email protected]>
> wrote:
>
> It seems that the modification only affects the weighing of slopes at the
> nodes between two flat regions. Adding a new parameter to the existing
> constructor with a backwards-compatible default value may be better than
> creating a whole new class.
>
> Option 1: original=True
> Option 2: modified=False
>
> It would certainly be nice to have both versions available in scipy.
>
> - Joe
>
> On Tue, Apr 11, 2023, 05:52 Edoardo Marconi <
> [email protected]> wrote:
>
>> Hello everyone,
>>
>> I was looking for a Scipy equivalent of MATLAB’s 'makima' function (see
>> https://it.mathworks.com/help/matlab/ref/makima.html). Apparently, it
>> does not exist yet, and I could not find any related posts in this mailing
>> list either.
>>
>> As Matlab provides a clear explanation of this modified version of
>> Akima's algorithm, it was straightforward to implement it starting from the
>> existing 'Akima1DInterpolator' class from scipy.interpolate.
>>
>> I'd love to see this integrated into SciPy, could a new
>> 'ModifiedAkima1DInterpolator' class be an interesting addition?
>> I have never contributed to SciPy and may need some guidance. Any
>> suggestions on how to proceed?
>>
>> Thanks in advance for any comments!
>> _______________________________________________
>> SciPy-Dev mailing list -- [email protected]
>> To unsubscribe send an email to [email protected]
>> https://mail.python.org/mailman3/lists/scipy-dev.python.org/
>> Member address: [email protected]
>>
> _______________________________________________
> SciPy-Dev mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
> https://mail.python.org/mailman3/lists/scipy-dev.python.org/
> Member address: [email protected]
>
>
> _______________________________________________
> SciPy-Dev mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
> https://mail.python.org/mailman3/lists/scipy-dev.python.org/
> Member address: [email protected]
>

_______________________________________________
SciPy-Dev mailing list -- [email protected]
To unsubscribe send an email to [email protected]
https://mail.python.org/mailman3/lists/scipy-dev.python.org/
Member address: [email protected]