PSA_Final_Project_Spring22

Code Repository for INFO 6205 Spring 22 Final Project

Team members:

Shashank Siripragada, Section 8 NUID: 002193773

Mayannk Kumaar, Section 8 NUID: 001537115

This project requires installation of python 3

Please install python 3

Testing

Before training, run all the test cases in tests.py using

python tests.py

Training

To train the MENACE algorithm, run

python game.py --mode train --iterations 1000 --probability 0.9

• mode : train (for training mode)
• iterations : Number of iterations to train MENACE
  • iterations default : 1000
• probability : probability that the human agent opponent follows the ideal strategy
  • probability default : 0.7 (probability = 1 - epsilon, where epsilon = 0.3)

Play against trained MENACE

To play against the trained MENACE, run

python game.py --mode play --ngames 2

• mode : play
  • for playing against MENACE
• ngames : Number of games to play against MENACE default : 1

This will initialize a command line interface enabling user to play against MENACE