Update/cleanup/standardize notebooks

[jGaboardi]

This is a relatively large PR that touches all notebooks. It cleans up, modernizes, and standardizes both notebook sets, region and locate, but mainly focuses on highlighting and consolidating the work by @gegen07 from GSoC 2021 and @erinrolson from GSoC 2022 (awesome work from both!)

• resolves update docs for new location models #269, LSCP Notebook Plots  #263
• (hopefully) resolves P-Median Problem Tutorial Binder example #276
• work toward Tutorial notebooks need more narrative #101
• xrefs short-to-mid term plan for improving docs, structure, and formatting #280, test spopt against the shapely==2.0b1 #289

Also overhauls spopt.locate.utils.simulate_geo_points() for robustness and adds tests.

[jGaboardi]

It also cleans up and adds the two notebooks to the docs that @TimMcCauley contributed (we forgot to add those in previously). Thanks again, @TimMcCauley!

[TimMcCauley]

It also cleans up and adds the two notebooks to the docs that @TimMcCauley contributed (we forgot to add those in previously). Thanks again, @TimMcCauley!

Happy to! If you have further ideas, I'd be happy to whip something 🆙

[jGaboardi]

@TimMcCauley I am happy to review anything you'd be willing to contribute!

[jGaboardi]

@gegen07 Any chance of getting a review of this PR for merge? Not trying to rush if you don't have the time. Also, there are no functionality changes in this PR, mostly just sprucing up of notebooks, etc.

*There are minor functionality changes in simulate_geo_points().

[gegen07]

I'll review it tomorrow. Sorry for delay, I've been quite busy this week.

[jGaboardi]

I'll review it tomorrow. Sorry for delay, I've been quite busy this week.

No problem! Thanks so much!

[jGaboardi]

Addressing #289 within this PR, as well (@martinfleis)

[jGaboardi]

OK, @martinfleis I've found something for you on the shapely==2.0b1 front. Here's a MWE to reproduce:

Python 3.10.6 | packaged by conda-forge | (main, Aug 22 2022, 20:41:54) [Clang 13.0.1 ]
Type 'copyright', 'credits' or 'license' for more information
IPython 8.5.0 -- An enhanced Interactive Python. Type '?' for help.

In [1]: import numpy
   ...: import shapely
   ...: from shapely.geometry import Point
   ...: shapely.__version__
Out[1]: '2.0b1'

In [2]: x, y = numpy.array([1]), numpy.array([1])

In [3]: x, y
Out[3]: (array([1]), array([1]))

In [4]: Point(x, y)
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In [4], line 1
----> 1 Point(x, y)

File ~/miniconda3/envs/py310_shapely_20b1/lib/python3.10/site-packages/shapely/geometry/point.py:77, in Point.__new__(self, *args)
     74     geom = shapely.points(np.array(args))
     76 if not isinstance(geom, Point):
---> 77     raise ValueError("Invalid values passed to Point constructor")
     78 return geom

ValueError: Invalid values passed to Point constructor

Two singular value numpy.array objects were previously valid inputs to construct a Point. Is this a regression in shapely or something that should change from the user's side?

[martinfleis]

@jGaboardi good catch! That'll be a regression on shapely side. I'll report it there.

[jGaboardi]

@gegen07 We can ignore the failing 310-DEV_shapely20b1 tests for now.

[jGaboardi]

ValueError: Invalid values passed to Point constructor being addressed in shapely/shapely#1590

[gegen07]

@jGaboardi LGTM!
Just one comment that I didn't get!