def main():
if len(sys.argv) >= 2 and sys.argv[1] == 'predict':
first = bitify(sys.argv[2])
second = bitify(sys,argv[3])
result = predict_mlp(first, second)
print(str(result))
sys.exit(0)
print("Gathering data in memory...")
dbn_data, mlp_data, mlp_labels = gather_data()
print("Gathering Data Done!")
print("Splitting data in Training and Test Data...")
X_train, X_test, Y_train, Y_test = train_test_split(
mlp_data,
mlp_labels,
test_size=0.25
)
print("Splitting Done!")
print("Training Deep Belief Network...")
train_dbn(dbn_data)
print("Training DBN Done!")
print("Training Multi-Layer Perceptron...")
train_mlp(X_train, Y_train)
print("Training MLP Done!")
print("Testing MLP...")
print(f"Score = {test_mlp(X_test, Y_test)}")
def eval(board, alpha, beta, player):
if ((player == chess.WHITE and alpha.turn == chess.WHITE and alpha.is_checkmate())
or (player == chess.BLACK and beta.turn == chess.BLACK and beta.is_checkmate())):
return board
if player == chess.WHITE and model.predict_mlp(utils.bitify(board.fen()), utils.bitify(alpha.fen()))[0] < 0.5:
return alpha
elif player == chess.BLACK and model.predict_mlp(utils.bitify(board.fen()), utils.bitify(beta.fen()))[0] < 0.5:
return beta
return board
def extract_mlp_data(data):
returned_data = list()
returned_labels = list()
for data_point in data:
first_position = bitify(data_point[0].split(';')[0])
second_position = bitify(data_point[1].split(';')[0])
returned_data.append((first_position, second_position))
returned_labels.append((1, 0))
returned_data.append((second_position, first_position))
returned_labels.append((0, 1))
return returned_data, returned_labels
def _hash_code(self, x):
"""
Returns the polynomial hash code for string x
:param x: string to hash
:return: hash code of x
"""
return sum(int(c) * self._z**k for k, c in enumerate(bitify(x)))
def alphabeta(board: chess.Board, depth: int, alpha: chess.Board, beta: chess.Board, player: chess.Color):
if depth == 0:
return eval(board, alpha, beta, player)
if player == chess.WHITE:
for move in board.generate_legal_moves():
new_board = board.copy()
new_board.push(move)
alpha = alphabeta(new_board, depth-1, alpha, beta, chess.BLACK)
if model.predict_mlp(utils.bitify(beta.fen()), utils.bitify(alpha.fen()))[0] == 1:
break
return alpha
else:
for move in board.generate_legal_moves():
new_board = board.copy()
new_board.push(move)
beta = alphabeta(new_board, depth-1, alpha, beta, chess.WHITE)
if model.predict_mlp(utils.bitify(beta.fen()), utils.bitify(alpha.fen()))[0] == 1:
break
return beta
def put(self, data, is_bits=False):
if not is_bits:
data = utils.bitify(data)
self.obuffer += data
def extract_dbn_data(white_wins, black_wins):
return np.concatenate((
[bitify(i.split(';')[0]) for i in white_wins],
[bitify(i.split(';')[0]) for i in black_wins]
))
def put(self, data, is_bits=False):
if not is_bits:
data = utils.bitify(data)
self.obuffer += data