NBA Games Ranked is a free service that tweets a ranking of the previous day's NBA games, without revealing the final scores. It's designed to help NBA fans in non-American time zones find and watch the most exciting games. Follow me on Twitter: @NBAGamesRanked.
First, we collect data from RapidAPI and calculate the score for each game. The score is based on the number of arbitrary parameters. This is happening in the nba_games_ranked.py script that runs in a cloud as a cron job. Tha ranking is saved in a csv file.
Then, another script tweet_ranking.py - also running as a cron job - reads the ranking from the csv file and tweets it on @NBAGamesRanked account.
A high level overview of the installation process:
- Set up a VM machine
- Install Python virtual environment (venv) and required packages (see
requirements.txt) - Set up a Twitter Developer account and get the necessary API keys
- Set up a RapidAPI account and get the necessary API keys
- Write keys to a
.envfile - Set up a cron job to run 'nba_games_ranked.py' and 'tweet_ranking.py' scripts. The latter should run after the former finishes, so give it a few minutes head start.
This project and its code are available for personal use only. Individuals may use it solely on their personal devices. No results, outputs, or derived works may be published, shared, or distributed on any platform or in any form. For any use beyond personal purposes, please contact manduk.ai.