James Gaboardi

Woah boy...

Not 100% sure why [this is complaining](https://github.com/pysal/momepy/actions/runs/3091088424/jobs/5000737392#step:6:163) so much. The message indicates the problem is with using `.iloc`, but we are using [`.loc` there](https://github.com/pysal/momepy/blob/main/momepy/elements.py#L281). I had thought using `.loc` was...

~~Probably in [`numpy.nanpercentile`](https://github.com/pysal/momepy/blob/c574bcfd8445dc8b5e54912cd8fe28ea10ff2cf5/momepy/diversity.py#L904-L923), correct? Or is it specifically in `pandas`?~~

Excellent digging.

The [`TestElements.test_get_network_ratio` failure](https://github.com/pysal/momepy/actions/runs/3101130362/jobs/5022175606#step:6:124) appears to be also due to the `geopandas.clip()` problem. `Tessellation()`->`_enclosed_tessellation()`->`_tess()`->`_morphological_tessellation()`->`geopandas.clip()`

xref https://github.com/pandas-dev/pandas/pull/48711

xref #437

tl;dr – I like it. 1. IMHO temporary code duplication is acceptable; just as long as we prune it down as soon as possible. 2. I think it is as...

@knaaptime Do you have some advice for this?

Looks like we never add `spopt` into [`environment.yml`](https://github.com/pysal/spopt/blob/main/environment.yml)...