Choropleth Colors not Displaying

[joshdelarosa1]

Like issue #92, the colors on the Choropleth are not displaying. What is odd is the quantiles are being set and the data.json file contains the valid values, but all of the shapes of the same color.

I'm doing something wrong, but can't figure out what it is. The data files are in the same directory, so the solution isn't the same as #92.

import folium
import pandas as pd

state_geo = r'gz_2010_us_040_00_20m.json'
population = r'd.csv'

state_data = pd.read_csv(population)

#Let Folium determine the scale
map = folium.Map(location=[48, -102], zoom_start=3)
map.geo_json(geo_path=state_geo, data=state_data, data_out='data.json',
                columns=['state', 'D001'],
                key_on='feature.NAME',
                fill_color='YlGn', fill_opacity=0.7, line_opacity=0.1,
                legend_name='population')
map.create_map(path='index.html')

[ocefpaf]

Hi @joshdelarosa1,

We can't debug this without the data and the folium version you are using.

[joshdelarosa1]

thanks @ocefpaf
'folium==0.1.6'
the geo data can be found here:
http://eric.clst.org/wiki/uploads/Stuff/gz_2010_us_040_00_20m.json

here is the cvs data
state,D001
Alabama,4779736
Alaska,710231
Arizona,6392017
Arkansas,2915918
California,37253956
Colorado,5029196
Connecticut,3574097
Delaware,897934
District of Columbia,601723
Florida,18801310
Georgia,9687653
Hawaii,1360301
Idaho,1567582
Illinois,12830632
Indiana,6483802
Iowa,3046355
Kansas,2853118
Kentucky,4339367
Louisiana,4533372
Maine,1328361
Maryland,5773552
Massachusetts,6547629
Michigan,9883640
Minnesota,5303925
Mississippi,2967297
Missouri,5988927
Montana,989415
Nebraska,1826341
Nevada,2700551
New Hampshire,1316470
New Jersey,8791894
New Mexico,2059179
New York,19378102
North Carolina,9535483
North Dakota,672591
Ohio,11536504
Oklahoma,3751351
Oregon,3831074
Pennsylvania,12702379
Rhode Island,1052567
South Carolina,4625364
South Dakota,814180
Tennessee,6346105
Texas,25145561
Utah,2763885
Vermont,625741
Virginia,8001024
Washington,6724540
West Virginia,1852994
Wisconsin,5686986
Wyoming,563626
Puerto Rico,3725789

here is what the map looks like

[ocefpaf]

The problem is in your JSON:

{ "type": "Feature", 
  "properties": { "GEO_ID": "0400000US04",
                        "STATE": "04",
                        "NAME": "Arizona",
                        "LSAD": "",
                        "CENSUSAREA": 113594.084000 }, "geometry":{...}...

You are using a key_on=feature.NAME and the NAME does not exist in Feature. NAME is inside properties. The docs do say that feature.properties.NAME should work, but I just tested that and it is broken. Your best option is to use this GeoJSON and adapt your DataFrame to be the two letter code for the states.

(Next time please use links to all files. It is time consuming to copy-and-paste to create a csv just to test it.)

[ocefpaf]

I am closing this and I opened another issue (#223) to address the key_on option.

[joshdelarosa1]

@ocefpaf the sort on my data was also wrong, but after that was fixed, worked like a charm. thanks!

[ocefpaf]

Glad to help.

Sorry for reopening this, but I have the same problem and I'm wondering about what is meant by "the sort on my data was wrong"? What do I need to sort? I cannot find that hint in the docs?

[ocefpaf]

@bioprog see #220 (comment)

BTW the key_on='feature.NAME' is properly documented in the docstring:

folium/folium/folium.py

Lines 191 to 194 in 9a7d65c

	key_on: string, default None
	Variable in the GeoJSON file to bind the data to. Must always
	start with 'feature' and be in JavaScript objection notation.
	Ex: 'feature.id' or 'feature.properties.statename'.

Hi,
yes, the key_on argument is properly documented in the source code and I think I have applied it the right way(?). Anyway, you referred me to your comment asking for data. So here's a small example where the color gradient doesn't apply:

test_data.csv

plz,anzahl
"50931",50
"50935",1
"50937",100

test_data.geojson

script.py

import folium
import pandas as pd

osm = folium.Map([50.929654, 6.923508], zoom_start=8)
osm

plz_data = pd.read_csv(r'test_data.csv')

osm.choropleth(geo_path=r'test_data.geojson', data=plz_data, fill_color='YlGn', fill_opacity=0.7, line_opacity=0.2, columns=['plz','anzahl'],key_on='feature.properties.postal_code')
osm
osm.save('choropleth.html')

EDIT: Sorry, I have no idea why the long geojson doesn't get wrapped with a scroll bar? The geojson comprises 3 postal code areas of germany in the city of cologne.

[ocefpaf]

EDIT: Sorry, I have no idea why the long geojson doesn't get wrapped with a scroll bar? The geojson comprises 3 postal code areas of germany in the city of cologne.

You can attach a file or send a gist like I did above. Don't paste data like that 😉

[ocefpaf]

Worked for me:

http://nbviewer.jupyter.org/gist/ocefpaf/103d19edd67cb9dad562008f97a556af

What version of folium do you have there?

In your link I see all 3 areas having the same color? But at least 50937 should display a very dark green, shouldn't it? Or is this a browser issue? I'm using FF 55.0.2

EDIT
Oh I forgot to answer your question:

>>> import folium
>>> folium.__version__
'0.3.0'

EDIT2

[ocefpaf]

My bad your dataframe had ints and the json is read as string, this should work:

http://nbviewer.jupyter.org/urls/gist.githubusercontent.com/ocefpaf/103d19edd67cb9dad562008f97a556af/raw/ba03e2ade2466624ab6eb1d851d8edd1a9055d34/chropleth.ipynb

We could try to coerce the keys to str. PRs welcomed 😄

I never did a "PR" / pull request but only because it was never necessary. Actually I'm curious how to do it so I will read a tutorial and then contribute to the project. I will do it in my spare time on the weekend, maybe earlier.

In the meantime I use this, which works (thanks for the hint, I got a very nice map now):

plz_data = pd.read_csv(r'test_data.csv')
plz_data['plz'] = plz_data['plz'].astype(str)