def time_decision():
game = PentagoGame()
arrays = [
[
[None, 1, None, None, None, None],
[None, None, 1, None, 0, None],
[None, 0, None, 1, None, 0],
[None, None, None, None, 1, None],
[None, None, None, 0, None, None],
[None, None, None, None, None, None],
],
[
[None, None, None, None, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
],
[
[ 0, 0, 0, 0, None, None],
[None, None, 1, None, None, None],
[None, None, 1, None, None, None],
[None, None, 1, None, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
],
[
[None, None, None, None, None, None],
[None, None, None, None, None, None],
[None, 1, None, None, None, None],
[ 0, 0, None, 0, 0, 0],
[None, None, 1, None, 1, None],
[None, None, 1, None, None, None],
],
]
times = []
for b in map(array_to_bin, arrays):
state = game.make_state(b=b)
t0 = time.clock()
alphabeta_cutoff_search(state, game, d=2)
t1 = time.clock()
d = t1 - t0
times.append(d)
print("%.3f" % d)
print('search took %.3fs.' % sum(times))
print("times = ", times)
def move(array, token):
b = array_to_bin(array)
state = game.make_state(b=b)
i, q, clockwise = alphabeta_cutoff_search(state, game, d=2)
coor = (i // 6, i % 6)
center = [(1, 1), (1, 4), (4, 1), (4, 4)][q]
move = (coor, center, clockwise)
print("token: %s | move: %s" % (state.num_moves % 2, move))
return move
def demo_play():
array = [[None] * 6 for _ in range(6)]
num_moves = 0
state = GameState(array=array, num_moves=num_moves, utility=None)
while True:
move = alphabeta_cutoff_search(state, game, d=3)
print("token: %s | move: %s" % (state.num_moves % 2, move))
game.display(state)
print()
# input(' ....continue?')
new_state = game.result(state, move)
state = new_state
def decision_test1():
array = [
[0, 0, 0, 0, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
]
state = game.make_state(array=array, num_moves=1)
move = alphabeta_cutoff_search(state, game, d=1)
print('move:', move)
assert move[0] == (0, 4)
print("passes decision_test1")
def decision_test2():
array = [
[0, None, None, None, None, None],
[None, 0, None, None, None, None],
[None, None, 0, None, None, None],
[None, None, None, 0, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
]
state = GameState(array=array, num_moves=1, utility=None)
move = alphabeta_cutoff_search(state, game, d=1)
print(move)
# assert move[0] == (4, 4)
print("passes decision_test2")
def decision_test2():
array = [
[ 0, None, None, None, None, None],
[None, 0, None, None, None, None],
[None, None, 0, None, None, None],
[None, None, None, 0, None, None],
[None, None, None, None, None, None],
[None, None, None, None, None, None],
]
b = array_to_bin(array)
state = game.make_state(b=b, num_moves=1)
move = alphabeta_cutoff_search(state, game, d=1)
print('move:', move)
assert move[0] == 28
print("passes decision_test2")
player = 2
computer = 1
# action = game.alphabeta_cutoff_search(state,game)
# state = game.result(state,action)
while not exit:
print(state)
if game.to_move(state) == player:
entry = input("Please insert your move: <row> <column> ")
if "exit" in entry:
exit == True
elif len(entry.split(" ")) == 2:
row = int(entry.split(" ")[0])
column = int(entry.split(" ")[1])
state = game.result(state, (player, row, column))
elif game.to_move(state) == computer:
action = games.alphabeta_cutoff_search(state, game, d=4)
state = game.result(state, action)
else:
print("Error")
sys.exit(0)
if game.terminal_test(state):
print(state)
result = game.utility(state, player)
if result == 1:
print("YOU WON")
elif result == -1:
print("YOU LOST")
else:
print("DRAW")
sys.exit(0)
def make_alphabeta_player(N=0):
""" Returns a player function that uses alpha_beta search to depth N """
if N == 0:
return lambda g, s: games.alphabeta_search(s, g)
else:
return lambda g, s: games.alphabeta_cutoff_search(s, g, d=N)
def move(array, token):
num_moves = token
state = GameState(array=array, num_moves=num_moves, utility=None)
move = alphabeta_cutoff_search(state, game, d=2)
print("token: %s | move: %s" % (state.num_moves % 2, move))
return move
def alphabeta_depth8(game, state):
return games.alphabeta_cutoff_search(state, game, d=8)
print(a_game) # A player is represented by a search function that takes a game instance and a state and returns a move. The game's methods (actions, result, utility and terminal_test) do the real work # # The aima code defines several search functions and some players based on them # In[34]: # minimax returns a move by using the minimax algorithm # to search all the way down to leaves to choose best move minimax = lambda g, s: games.minimax_decision(s, g) # these return a move by using the alphabeta algorithm # to search to a given depth to choose best move dumb = lambda g, s: games.alphabeta_cutoff_search(s, g, d=1) smart = lambda g, s: games.alphabeta_cutoff_search(s, g, d=3) smarter = lambda g, s: games.alphabeta_cutoff_search(s, g, d=20) # Smartest return a move by using the alphabeta algorithm # to search down to leaves to choose best move smartest = games.alphabeta_player # random just returns a random move random = games.random_player # In[35]: a_game = TicTacToe() a_game.play_game(minimax, minimax)