bazingayu / Reinforcement_Learning_Tabular_Q_Learning_vs_minimax

Reinforcement Learning : Tabular Q learning Vs minimax to play tic tac toe and connect 4.

python play.py -g tic -p1 minimax -p2 human -iter 10
-g means the game you choice to play. (you can choose "tic" or "connect")
-p1 for player1, which algorithm do you want to use.
please make sure you have “minimax” for minimax player
you have "tq" for tabular Q-learning player
you have "human" for human player
otherwise, it will be default opponent.
-p2 means which algorithm do you want to use for player 2
-iter means how many iterations do you want to run. Each iteration contains 100 games to calculate the winning percentage.

If you have any question. Feel free to email [e-mail]