NBA Data Vis

This repository contains the code for the NBA game data visualizer project. The repository contains several main sub-directories, each of which is described below:

Server

The server/ directory contains code to setup, load data into, and run the server. The server serves NBA player/season/play-by-play data in various forms to potentially multiple clients. To run the server, run:

nodejs server.js

Make sure you node 6 or greater.

Before you start up the server, you'll want to make sure that the server database is populated with NBA data. You should use the load_data.js script to load the data into the database. The data itself can be fetched and post-processed using the scripts in the scripts/ directory. See the related section below.

Clients

This directory contains multiple different clients that utilize the data served by the server. At the moment, there are two clients:

dev-sandbox

As the name suggests, this client was developed as a sandbox to experiment with ways that the data from the server can be fetched and visualized. To use the client, just run a standard http web server from clients/dev-sandbox.

ballers

This client uses TSN-e plotting algorithms to plot the similarity between NBA players based on their statistics curves throughout the time of a game. To use the client, just run a standard http web server from clients/dev-sandbox.

Scripts

The scripts directory contains several shell and python scripts for downloading and processing the NBA data.

• fetch-nba-data.sh Is responsible for downloading and organizing the data from the web.
• process-data.py Does post-processing on the downloaded data, making it more friendly to use in an application
• setup-postgres-db.sh Was written to load the post-processed data into a postgres database. WARNING: I was going to make the server get the info from a postgres DB, but later changed my mind to use mongodb. This script does NOT NEED TO BE USED.
• do-all.sh Runs all of the above, in sequence.