utils.py

import numpy as np
from MCTS import MCTS
from games import TicTacToe

## Random vs Random
def play_game_random(game,player,state):
    while True:
        valid_moves=game.get_valid_moves(state)
        if np.sum(valid_moves)==0:
            return 0
        action=np.random.choice(np.where(valid_moves==1)[0])
        state=game.get_next_state(state,action,player)
        value,terminated=game.get_value_and_terminated(state,action)
        if terminated:
            if value==1:
                return player
            else:
                return 0
        player=game.get_opponent(player)

## MCTS vs Random
def play_game_mcts(game,mcts,player,state):
    while True:
        if player==1:   
            valid_moves=game.get_valid_moves(state)
            if np.sum(valid_moves)==0:
                return 0
            action=np.random.choice(np.where(valid_moves==1)[0])
        else:
            action_probs=mcts.search(state)
            action=np.argmax(action_probs)
        state=game.get_next_state(state,action,player)
        value,terminated=game.get_value_and_terminated(state,action)
        if terminated:
            if value==1:
                print(f"Player {player} wins")
                return player
            else:
                print("Draw")
                return 0
        player=game.get_opponent(player)