Convert a pandas DataFrame to GeoJSON (Stations)

In this Python snippet we use stations (geographic) from the Swiss public transportation and convert the data to a GeoJSON (http://geojson.org/) file.

Donwload the 'bfkoordgeo.csv' file, https://opentransportdata.swiss/en/dataset/bhlist/resource/b92a372f-7843-4ddd-b1c6-c9c6397e1097
Convert the file from ISO-8859-1 to UTF-8

iconv -f ISO-8859-1 -t UTF-8 bfkoordgeo.csv > out.csv

Note: The data is from the Open Data Platform Swiss Public Transport, https://opentransportdata.swiss/en/

Required libraries

pandas, http://pandas.pydata.org/
geojson, https://pypi.python.org/pypi/geojson/

Load the data

First let's load the data with pandas. The data frame contains the stations from the public transportations from Switzerland and some from adjoining countries. We have the columns:

StationID
Longitude
Latitude
Height
Remark

Longitude and Latitude should be WGS 84 coordinates.



In [2]:

    
import pandas as pd

df = pd.read_csv('data/bfkoordgeo_utf8.csv')
df.head()









    Out[2]:







  
    
      
      StationID
      Longitude
      Latitude
      Height
      Remark
    
  
  
    
      0
      6.0
      7.549783
      47.216111
      441.0
      St. Katharinen
    
    
      1
      7.0
      9.733756
      46.922368
      744.0
      Fideris
    
    
      2
      11.0
      7.389462
      47.191804
      467.0
      Grenchen Nord
    
    
      3
      16.0
      6.513937
      46.659019
      499.0
      La Sarraz, Couronne
    
    
      4
      22.0
      7.589551
      47.547405
      277.0
      Basel SBB

Now we do some data cleaning and remove all rows where Longitude and Latitude are 'null'.



In [3]:

    
df = df[df['Longitude'].notnull()]
df = df[df['Latitude'].notnull()]

# will display all rows that have null values
#df[df.isnull().any(axis=1)]

Convert pandas data frame to GeoJSON

Next we convert the panda data frame to geosjon objects (FeatureCollection/Feature/Point).



In [4]:

    
import geojson as geojson

values = zip(df['Longitude'], df['Latitude'], df['Remark'])
points = [geojson.Feature(geometry=geojson.Point((v[0], v[1])), properties={'name': v[2]}) for v in values]

geo_collection = geojson.FeatureCollection(points)

print(points[0])









    



{"geometry": {"coordinates": [7.549783000000001, 47.216111], "type": "Point"}, "properties": {"name": "St. Katharinen"}, "type": "Feature"}

Save the GeoJSON (FeatureCollection) to a file

Finally we dump the GeoJSON objects to a file.



In [5]:

    
dump = geojson.dumps(geo_collection, sort_keys=True)

'''
with open('stations.geojson', 'w') as file:
    file.write(dump)
'''









    Out[5]:





"\nwith open('stations.geojson', 'w') as file:\n    file.write(dump)\n"

Result

You can find the resulat (GeoJSON file) from this snippet here

stations.geojson



In [ ]:

	StationID	Longitude	Latitude	Height	Remark
0	6.0	7.549783	47.216111	441.0	St. Katharinen
1	7.0	9.733756	46.922368	744.0	Fideris
2	11.0	7.389462	47.191804	467.0	Grenchen Nord
3	16.0	6.513937	46.659019	499.0	La Sarraz, Couronne
4	22.0	7.589551	47.547405	277.0	Basel SBB