ValueError: Columns must be same length as key

Question

I want to geocode a number of universities in Europe (stored in a csv file). The csv file contains only one column (Name). However, I am yet to complete the work, since my code contains bugs. When I ran the code, I got ValueError: Columns must be same length as key. This occurred on the last line. How do I solve this?

Here's a sample of the code I used:

from geopy.geocoders import Nominatim

import pandas as pd

import numpy as np

geolocator = Nominatim(user_agent = 'ed') 

universities = pd.read_csv("Z:/data/top_universities.csv")

def my_geocoder(row):
    try:
        point = geolocator.geocode(row).point
        return pd.Series({'Latitude': point.latitude, 'Longitude': point.longitude})
    except:
        return None

universities[['Latitude', 'Longitude']] = universities.apply(lambda x: my_geocoder(x['Name']), axis=1)

Answer 1

pandas is expecting two values (columns) to be returned as you specified two keys ['Latitude', 'Longitude'].

However, when there's an exception, e.g. no geocoding result, you only return a single value None. So you get ValueError: Columns must be same length as key

Return two values when there's a geocoding error, i.e. None, None and you can filter them out later:

from geopy.geocoders import Nominatim
import pandas as pd
import numpy as np


def my_geocoder(row):
    try:
        point = geolocator.geocode(row).point
        return pd.Series({'Latitude': point.latitude, 'Longitude': point.longitude})
    except:
        return None, None


geolocator = Nominatim(user_agent = 'ed')
universities = pd.read_csv("test.csv")
universities[['Latitude', 'Longitude']] = universities.apply(lambda x: my_geocoder(x['Name']), axis=1)

print(universities.head())

Output:

                             Name   Latitude   Longitude
0  Australian National University -35.281213  149.116779
1            University of Sydney -33.896083  151.184661
2                            RMIT -37.682204  145.068329
3              No Such University        NaN         NaN