ML_Project

Repository for Notre Dame Machine Learning project

Python libraries needed to run code

All of the code for this project was run using Python3.6

• pandas
• os
• sys
• collections
• math
• seaborn
• matplotlib
• itertools
• xgboost
• sklearn
• warnings
• string
• pickle
• numpy

Running

To run the code, simply call the following command in the terminal

./run_all.sh

This code runs the following commands in order

python3.6 parse_data.py
python3.6 generate_matchups.py
python3.6 get_matchup_results.py
python3.6 algorithms_seth.py
python3.6 krieg_mm_analysis.py
python3.6 krieg_plot_mm_results.py

If you would like to use the new parameter GridSearch results from algorithms_seth.py, they must be replaced in the model parameters in krieg_mm_analysis.py by hand.