Abyssalearn

Database

A copy of database.db is maintained in data/database.db, please avoid committing any of your changes to this file.

You can discard your local changes with git reset --hard HEAD before pulling. N.B. Make sure you have committed your own changes before running this command as it will wipe out any uncommitted changes from the filesystem!

Installation

  1. Install pipenv from your distribution package manager or with pip install pipenv
  2. Run pipenv install --ignore-pipfile to install dependencies

Usage

Run pipenv shell and then use any of the following features:

  • python main.py scrape - Scrapes any new contract data from the EVE servers
  • python main.py summary - Print a summary of the data in the database
  • python main.py random 47702 - Pull a random module with type ID 47702 (abyssal stasis webifier) and print its pricing/attributes
  • python main.py ridgereg|lassoreg 47702 2 0.001 - Run Ridge|Lasso Regression with type ID 47702, features of degree 2 and alpha=0.001 and print out the mean squared error of 5-Fold cross-validation
  • python main.py linreg 47702 - Run Linear Regression with type ID 47702 and print out the mean squared error of 5-Fold cross-validation
  • python main.py ridgereg-crossval|lassoreg-crossval 47702 2 - Run Ridge|Lasso Regression with type ID 47702 and features of degree 2, and test alpha in [0.0001, 0.001, 0.01, 0.1, 1] using 5-Fold cross-validation
  • python main.py polyreg 47702 2 - Run Linear Regression with type ID 47702 and polynomial features with degrees 2 print out the mean squared error of 5-Fold cross-validation
  • python main.py polyreg-crossval 47702 - Run Linear Regression with type ID 47702 and polynomial features with degrees in ranges [1,2,3,4,5,6] and print out the mean squared error of 5-Fold cross-validation