A look into recent traffic accident information

Dataset provided by Kaggle on the following link: https://www.kaggle.com/sobhanmoosavi/us-accidents

The ETL process is governed by the script in /python/ingest_crashes.py.

To update the crashes database located at /db/crahses.db, ensure the target source data is in the /csvfiles/traffic_crashes.csv file and then run this script. It will report on how many values are added for each column.

This ETL only needs to be run if the contents of /csvfiles/traffic_crashes.csv changes.

• Python 3.7
• Traffic Crash data in /csvfiles/traffic_crashes.csv
• Pandas (pip install pandas)
• SQLite (pip install sqlite3)

A development version of the REST API is available in /python/publish_crashes.py. Run this script to set up a dev web server with the following 3 routes:

• /: Redirects to /refresh the first time this is run; to /crashes otherwise
• /crashes: Returns crashes as a JSON object
• /refresh: Refreshes the crashes from the database before redirecting to /crashes

We should look into a way to more-durably publish this before presenting, but this will get us started with our vizzes.

• Python 3.7
• A /db/crashes.db file with at least an empty crash table.
  • See the ETL section above for details
• SQLAlchemy (pip install SQLAlchemy)
• SQLite (pip install sqlite3)
• Flask CORS (pip install flask-cors)