Skip to content

JakeUrban/mlhoops

Repository files navigation

MLHoops

Data collection and analysis application for NCAA basketball game predictions. MLHoops consists of a web scraping system, object relational mapping layer, and a collection of models built in a variety of languages and libraries. The web scraping system has collected information on every player, team, and game for every NCAA season since 1993.

Install

Installing according to these instructions will set up the development environment needed to use this package. It will NOT install a copy of the dataset I have collected by using this package.

Dependencies:

  • python3
  • pip3
  • mysql

Install the environment:

  • virtual environment
    • $ make venv

Initialize the database:

  • Create a local mysql database <db_name> (mlhoops is suggested)
  • Activate the virtual environment, then:
    • $ export SQLALCHEMY_DATABASE_URI='mysql+pymysql://<your_user>@localhost/<db_name>'
    • $ python3 mlhoops/mlhoops/db/init_db.py
      • This will create the database schema and add sample rows to each table for inital testing.

About

Machine learning application for NCAA basketball game predictions

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published