BUG: explore() using cmap instead of color attribute for *list-like of colors* when if column and categorical is set

[turbotimon]

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of geopandas.

• (optional) I have confirmed this bug exists on the main branch of geopandas.

---

Note: Please read this guide detailing how to provide the necessary information for us to reproduce your bug.

Code Sample, a copy-pastable example

import geopandas as gpd
data = {"type": "FeatureCollection", "features": [{"id": "1006", "type": "Feature", "properties": {"cluster": 1}, "geometry": {"type": "Point", "coordinates": [8.20894, 47.48112]}}, {"id": "1012", "type": "Feature", "properties": {"cluster": 0}, "geometry": {"type": "Point", "coordinates": [9.43357, 47.51026]}}, {"id": "1019", "type": "Feature", "properties": {"cluster": 1}, "geometry": {"type": "Point", "coordinates": [7.5748328, 47.5686935]}}, {"id": "1024", "type": "Feature", "properties": {"cluster": 2}, "geometry": {"type": "Point", "coordinates": [9.03017, 46.1963]}}, {"id": "1025", "type": "Feature", "properties": {"cluster": 2}, "geometry": {"type": "Point", "coordinates": [7.99028, 46.31906]}}]}
gdf = gpd.GeoDataFrame.from_features(data["features"])

# Plot
gdf.explore(
    column="cluster",
    categorical=True,
    legend=True,
    color=["red", "green", "blue"], # does not work, despite docs say so
    # cmap=["red", "green", "blue"], # works, but not in the docs
)

Problem description

I was trying to set colors for a categorical plot but was not able doing so with colors (list-like of colors) as the docs say.
Instead, after inspecting the source code, I saw that it is possible passing the list as cmap. BUT, this is not in the docs (only says str, matplotlib.Colormap, branca.colormap or function)

Recomendation

• Either: being able to pass list-like of colors with colors
• Or: Adding to the docs for cmap: str, matplotlib.Colormap, branca.colormap, function or array-like

Happy to do a MR for both (whatever you choose). However, in my opinion colors would be more consistent and intuitive.

Output of geopandas.show_versions()

SYSTEM INFO

python : 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]
executable : /usr/bin/python3
machine : Linux-6.1.58+-x86_64-with-glibc2.35

GEOS, GDAL, PROJ INFO

GEOS : 3.11.2
GEOS lib : None
GDAL : 3.6.4
GDAL data dir: /usr/local/lib/python3.10/dist-packages/fiona/gdal_data
PROJ : 9.3.0
PROJ data dir: /usr/local/lib/python3.10/dist-packages/pyproj/proj_dir/share/proj

PYTHON DEPENDENCIES

geopandas : 0.13.2
numpy : 1.25.2
pandas : 1.5.3
pyproj : 3.6.1
shapely : 2.0.2
fiona : 1.9.5
geoalchemy2: None
geopy : 2.3.0
matplotlib : 3.7.1
mapclassify: 2.6.1
pygeos : None
pyogrio : None
psycopg2 : 2.9.9 (dt dec pq3 ext lo64)
pyarrow : 10.0.1
rtree : None

[m-richards]

Thanks for reporting this @turbotimon, I'll let @martinfleis suggest what the best option is in terms of merging into one keyword or supporting both properly as currently documented.

[martinfleis]

cmap works by accident and color does not work together with column. It is one or the other, we should probably document that better. We also should support mapping of colours to values in some way but that hasn't been implemented yet. It is part of the plotting refactoring project that is on the roadmap for 1.x release once 1.0 is out.

[turbotimon]

color does not work together with column

I hope this is not intentional and also included in the plotting refactoring? As in my copy-paste example, it is very helpful to pass a list of colors along with a categorical column to give the categories a meaningful or more visible color compared to the map background.

Anyway, let me know if I can help with something.

[martinfleis]

column should be used together with cmap to assign colours to values. color should be used alone to simply assign a color to a geometry. Not ideal, I'm aware.

The refactoring is not planned in detail yet but this use case will surely be included.