`plot` doesn't work for pooled classifications

[darribas]

The plot utility function for classification objects does not seem to work for pooled classifications. This might be expected behaviour or not supported functionality but just in case.

Reproducible error:

from pysal.lib import examples
import mapclassify
import geopandas


mx_ex = examples.load_example('mexico')
mx = geopandas.read_file(mx_ex.get_file_list()[0])

years = ['PCGDP1940', 'PCGDP1960', 'PCGDP1980', 'PCGDP2000']
pooled = mapclassify.Pooled(mx[years])

for i, y in enumerate(years):
    classi = pooled.col_classifiers[i]
    classi.plot(mx)

which on the gds_env:6.1 returns:

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
/opt/conda/lib/python3.8/site-packages/geopandas/plotting.py in _mapclassify_choro(values, scheme, **classification_kwds)
   1010     try:
-> 1011         scheme_class = schemes[scheme]
   1012     except KeyError:

KeyError: 'pooled quantiles'

During handling of the above exception, another exception occurred:

KeyError                                  Traceback (most recent call last)
/opt/conda/lib/python3.8/site-packages/geopandas/plotting.py in _mapclassify_choro(values, scheme, **classification_kwds)
   1014         try:
-> 1015             scheme_class = schemes[scheme]
   1016         except KeyError:

KeyError: 'pooled quantiles'

During handling of the above exception, another exception occurred:

ValueError                                Traceback (most recent call last)
<ipython-input-20-f10e01e9c42e> in <module>
     12 for i, y in enumerate(years):
     13     classi = pooled.col_classifiers[i]
---> 14     classi.plot(mx)

/opt/conda/lib/python3.8/site-packages/mapclassify/classifiers.py in plot(self, gdf, border_color, border_width, title, legend, cmap, axis_on, legend_kwds, file_name, dpi, ax)
   2336             fmt = legend_kwds.pop("fmt")
   2337 
-> 2338         ax = gdf.assign(_cl=self.y).plot(
   2339             column="_cl",
   2340             ax=ax,

/opt/conda/lib/python3.8/site-packages/geopandas/plotting.py in __call__(self, *args, **kwargs)
    923             kind = kwargs.pop("kind", "geo")
    924             if kind == "geo":
--> 925                 return plot_dataframe(data, *args, **kwargs)
    926             if kind in self._pandas_kinds:
    927                 # Access pandas plots

/opt/conda/lib/python3.8/site-packages/geopandas/plotting.py in plot_dataframe(df, column, cmap, color, ax, cax, categorical, legend, scheme, k, vmin, vmax, markersize, figsize, legend_kwds, categories, classification_kwds, missing_kwds, aspect, **style_kwds)
    750             classification_kwds["k"] = k
    751 
--> 752         binning = _mapclassify_choro(values[~nan_idx], scheme, **classification_kwds)
    753         # set categorical to True for creating the legend
    754         categorical = True

/opt/conda/lib/python3.8/site-packages/geopandas/plotting.py in _mapclassify_choro(values, scheme, **classification_kwds)
   1015             scheme_class = schemes[scheme]
   1016         except KeyError:
-> 1017             raise ValueError(
   1018                 "Invalid scheme. Scheme must be in the set: %r" % schemes.keys()
   1019             )

ValueError: Invalid scheme. Scheme must be in the set: dict_keys(['boxplot', 'equalinterval', 'fisherjenks', 'fisherjenkssampled', 'headtailbreaks', 'jenkscaspall', 'jenkscaspallforced', 'jenkscaspallsampled', 'maxp', 'maximumbreaks', 'naturalbreaks', 'quantiles', 'percentiles', 'stdmean', 'userdefined'])

[sjsrey]

This is because classi in classi = pooled.col_classifiers[i] is an instance of a classifier - in other words, it has already been fit.

To see this, try to change the loop as follows:

from pysal.lib import examples
import mapclassify
import geopandas


mx_ex = examples.load_example('mexico')
mx = geopandas.read_file(mx_ex.get_file_list()[0])

years = ['PCGDP1940', 'PCGDP1960', 'PCGDP1980', 'PCGDP2000']
pooled = mapclassify.Pooled(mx[years])

for i, y in enumerate(years):
    classi = pooled.col_classifiers[i]
    label = f'pooled-{y}'
    mx[label] = classi.yb
    mx.plot(column=label, categorical=True, legend=True)

[sjsrey]

I think this is no longer an issue given the book chapter?