main.py

from functools import partial
import logging
import numpy as np
import random
import sys

from gamestate import GameState
from minimax import Minimax
from montecarlotreesearch import MonteCarloTreeSearch
from player import Player


def partial_mcts(number_of_simulation, game_state):
    return MonteCarloTreeSearch.get_best_move(game_state, number_of_simulation)


if __name__ == '__main__':
    # https://docs.python.org/3/library/logging.html#levels
    logging.basicConfig(format='%(message)s', stream=sys.stdout, level=logging.INFO)

    simulation_count = 100
    p1 = Player('P1', partial(partial_mcts, simulation_count))
    p2 = Player('P2', lambda game_state: Minimax.get_best_move(game_state))
    # p2 = Player('P2', lambda game_state: random.choice(game_state.get_valid_moves()))
    players = [p1, p2]

    board_size = 3
    game = GameState(np.array([p1, p2]), turn=1, game_board=np.zeros((board_size, board_size)))
    logging.debug(game)

    i = 0
    while not game.is_game_over:
        player = players[i % 2]
        logging.info(f'\n{player}\'s turn')
        game = game.make_move(player.get_move(game))
        print(game)
        i = i + 1

    print(f'\n{game.winner} won in {i} turns.')