Sergio Rey

It started failing after [deps added to ci](https://github.com/pysal/tobler/commit/704d7dddfa8f304b6d1bae11aeb0c8f96c3e9128#diff-515324956ca2a236d7ddd9b922b7323ac9a96e0332cce7e9586ee105316fd3b0R32) Since these are now in the [pyproject.toml](https://github.com/pysal/tobler/blob/d87ed4dd26fad4ef528995a9c2f607d052177e74/pyproject.toml#L51), maybe we can remove them from the ci?

+1 on moving the vba classification into `mapclassify`

One thought is we may want to consider a federation-wide policy on this. I'm old enough to remember that we started before there was a thing called pandas (or numpy...

> I am still working on this, but I recall now why the implementation of an "alternative" argument was a bit trickier than I expected... because we allow for the...

> OK, this should be ready to review in full/merge. The example notebook is complete. > > This commits the estimators in the same convention as the existing `Moran` estimators,...

For the 25.7 meta release, as an example, this would mean the minimum dependencies would be: - python 3.11 - numpy 1.26 - scipy 1.12 - matplotlib 3.80 - pandas...

Since the cadence reported on https://scientific-python.org/specs/spec-0000/ is quarterly, if we adopt a standard `pyproject.toml` across the ecosystem, we could construct such tooling to check for compliance.

@martirenom thanks for this. Would you be able to add a test suite and documentation so that users have an understanding of the scope of this option and what the...

> I like that automatic comparison but in some cases, it may be worth adding a note explaining _why_ there's no match and what is recommended to do instead. E.g...

This is still WIP. I have split the migration material and member comparisons apart for different audiences (users,devs). This also reorganizes the notebooks and changes the tutorial to a user-guide....