Geocoding Tools

Geocoding in R

The R package tidygeocoder provides a friendly interface to the US Census Bureau Public Geocoder API.

This repository contains helper scripts (rscripts/geocode.r) that help with batch processing of large datasets using functions from the tidygeocoder package.

Address validation in Python

The Python package usps-api provides a friendly interface to the US Postal Service Web Tools API. This API includes an address validation service that can be used to clean small numbers of addresses.

This repository contains helper scripts (geocodetoools/address_validator.py) that help with processing of addresses requiring validation.

Geocoding in Python

This repository also contains Python wrappers to tidygeocoder scripts using the rpy2 library for running R embedded in a Python process.

These wrappers allow the address validation and geocoding workflows to be unified in Python; for example, through the creation of a single Jupyer notebook or Python script that performs both the R-native and Python-native tasks.

Quickstart

Clone the repository

git clone https://github.com/kellypierce/geocode-tools.git

Environment setup

R requirements

• packages listed in renv.lock

Python requirements

• packages listed in requirements.txt
• install locally using the provided setup.py script (e.g. pip3 install . from within the repository directory)

USPS Web Tools API key

The Python scrips in this repository are designed to use the python-dotenv library to store and retrieve credentials saved as environment variables in a .env file.

• Register for API credentials at https://www.usps.com/business/web-tools-apis/
• Create a .env file in the repository directory and assign the variables USPS_API_UID and USPS_API_PWD the values of your user ID and password, respectively.