[SciPy-Dev] SciPy-Dev Digest, Vol 178, Issue 18

[Matt Newville]

Hi All,


On Tue, Aug 21, 2018 at 10:01 PM <scipy-dev-request at python.org> wrote:

> Send SciPy-Dev mailing list submissions to
>         scipy-dev at python.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         https://mail.python.org/mailman/listinfo/scipy-dev
> or, via email, send a message with subject or body 'help' to
>         scipy-dev-request at python.org
>
> You can reach the person managing the list at
>         scipy-dev-owner at python.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of SciPy-Dev digest..."
>
>
> Today's Topics:
>
>    1. NumPy 1.15.1 released. (Charles R Harris)
>    2. Proposal for some new fitting routines (Joseph Fox-Rabinovitz)
>    3. Re: Proposal for some new fitting routines (josef.pktd at gmail.com)
>
>
> Message: 2
> Date: Tue, 21 Aug 2018 22:12:49 -0400
> From: Joseph Fox-Rabinovitz <jfoxrabinovitz at gmail.com>
> To: SciPy Developers List <scipy-dev at python.org>
> Subject: [SciPy-Dev] Proposal for some new fitting routines
> Message-ID:
>         <
> CAAa1KPZ76rECAtBNF_cNMgpQD2eE1YiZgo9notpUSpGA-_yO6Q at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> I had sent out an email about a PR proposing an exponential and a power fit
> function, but I did not specify the PR number. The PR is #9158.
>
> Regards,
>
> - Joe
>
>
> On Mon, Aug 20, 2018, 3:40 PM Joseph Fox-Rabinovitz <
> jfoxrabinovitz at gmail.com> wrote:
>
> > I have added a function for fitting exponential and power curves. This
> > seems to
> > be a fairly common problem (e.g., see the list of selections from stack
> > exchange
> > below). The algorithm I have implemented here is objectively better than
> > the
> > common solution of fitting to log(y) because it requires no up-front
> > knowledge
> > of the intercept, or any other estimations.
> >
> > Assuming that this PR is acceptable, I would like to get some criticism
> on
> > where
> > I chose to put it in the directory structure. My thought was to create a
> > separate
> > module for such common non-iterative optimizations.
> >
> > I am also unsure of whether my citation of the original work is formatted
> > properly. I would like to make sure that the author gets the credit and
> > publicity they
> > deserve. The paper is mentioned (by the author) in answers to [1] and [2]
> > below.
> >
>

My impression is that these are fine procedures to have readily available,
but that they might not be a great fit for scipy.optimize since the basic
algorithm is non-iterative.  That is, the procedures seems closer to
scipy.stats.linregress and so it might make more sense having them
scipy.stats, perhaps calling the routines scipy.stats.expregress and
.powregress.  It would be very helpful to make the output a closer match to
those of linregress, especially including the standard errors and
correlation coefficients if possible.

Since the algorithms are non-iterative, I'm not sure it makes a lot of
sense to have these become the exponential and power-law fit methodsin
lmfit (there are existing exponential and power-law models there already),
but I would be inclined to use these algorithms (whether they go into scipy
or not) to produce initial guesses for these models.

Hope that helps (and thanks for asking!),

--Matt Newville
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scipy-dev/attachments/20180822/980c0fb8/attachment.html>