|
8 years ago | |
---|---|---|
API.json | 8 years ago | |
README.md | 8 years ago | |
cities.csv | 8 years ago | |
cities.txt | 8 years ago | |
console_output.txt | 8 years ago | |
geocode-vs.py | 8 years ago | |
outliers.txt | 8 years ago |
Getting a unified way to access the most important geocoding APIs (Nominatim, Google, MapQuest, HERE, OpenCage) to compare and evaluate.
Just run the script by providing a .txt file with each location to geocode on a new line, e.g.:
cat places.txt
Berlin
Amsterdam
Montpellier
...
python geocode-vs.py places.txt
The script will access API.json
for API keys (if neccessary) and will output cities.csv
as well as outliers.txt
. Don't forget to update your API.json
to get access to the MapQuest, HERE and OpenCage geocoding results. The outliers.txt
is some kind of shoot from the hip for evaluating the quality of the data. Outliers will be summarized in the console and simply mean that the coordinates for a location are not within the standard deviation of all 5 providers. This is by no means a real measure for quality, just a hint or a reminder to implement some real quality measurement.
API.json
- credentials for the geocoding APIs
cities.csv
- main output; CSV with header; example results included
cities.txt
- input file; each location on new line; examples included
console_output.txt
- example for console output
outliers.txt
- supposed outliers
ToDo: