This repository contains scripts to work with publicly available data for Citi Bikes in New York City. There is an associated blog post here.

Fetch available data files for individual rides and combine them into a single file rides.rds. If the file is already present, only new data is fetched and appended. Also fetches and saves station information from the JSON feed.

Visualize how bikes flow between different parts of the city during the day.

K-means clustering based on the average number of bikes arriving to and leaving from each station every hour, calculated separately for weekdays and weekends.