class GtpConnection():
def __init__(self, go_engine, debug_mode = False):
"""
Play Go over a GTP connection
Parameters
----------
go_engine: GoPlayer
a program that is capable of playing go by reading GTP commands
komi : float
komi used for the current game
board: GoBoard
SIZExSIZE array representing the current board state
"""
self.stdout = sys.stdout
self._debug_mode = debug_mode
sys.stdout = self
self.go_engine = go_engine
self.go_engine.komi = 6.5
self.go_engine.selfatari = 1
self.go_engine.pattern = 1
self.board = GoBoard(7)
self.mm_file_name = "features_mm_training.dat"
self.init_mm_file = False
self.num_game = 0
self.skip_counter = 0
self.param_options = {
"selfatari" : self.go_engine.selfatari,
"pattern" : self.go_engine.pattern
}
self.commands = {
"protocol_version": self.protocol_version_cmd,
"quit": self.quit_cmd,
"name": self.name_cmd,
"boardsize": self.boardsize_cmd,
"showboard": self.showboard_cmd,
"clear_board": self.clear_board_cmd,
"komi": self.komi_cmd,
"version": self.version_cmd,
"known_command": self.known_command_cmd,
"set_free_handicap": self.set_free_handicap,
"genmove": self.genmove_cmd,
"list_commands": self.list_commands_cmd,
"play": self.play_cmd,
"final_score": self.final_score_cmd,
"legal_moves": self.legal_moves_cmd,
"policy_moves": self.policy_moves_cmd,
"random_moves": self.random_moves_cmd,
"go_param": self.go_param_cmd,
"gogui-analyze_commands": self.gogui_analyze_cmd,
"num_sim": self.num_sim_cmd,
"showoptions": self.showoptions_cmd,
"feature_move": self.feature_move_cmd,
"features_mm_file": self.feature_mm_cmd
}
# used for argument checking
# values: (required number or arguments, error message on argnum failure)
self.argmap = {
"boardsize": (1, 'Usage: boardsize INT'),
"komi": (1, 'Usage: komi FLOAT'),
"known_command": (1, 'Usage: known_command CMD_NAME'),
"set_free_handicap": (1, 'Usage: set_free_handicap MOVE (e.g. A4)'),
"genmove": (1, 'Usage: genmove {w, b}'),
"play": (2, 'Usage: play {b, w} MOVE'),
"legal_moves": (0, 'Usage: legal_moves does not have arguments'),
"go_param": (2,'Usage: goparam {{{0}}} {{0,1}}'.format(' '.join(list(self.param_options.keys())))),
"num_sim":(1,'Usage: num_sim #(e.g. num_sim 100 )'),
"showoptions":(0,'Usage: showoptions does not have arguments'),
"feature_move":(1,'Usage: feature_move move')
}
def __del__(self):
sys.stdout = self.stdout
def write(self, data):
self.stdout.write(data)
def flush(self):
self.stdout.flush()
def start_connection(self):
"""
start a GTP connection. This function is what continuously monitors
the user's input of commands.
"""
self.debug_msg("Start up successful...\n\n")
line = sys.stdin.readline()
while line:
self.get_cmd(line)
line = sys.stdin.readline()
def get_cmd(self, command):
"""
parse the command and execute it
Arguments
---------
command : str
the raw command to parse/execute
"""
if len(command.strip(' \r\t')) == 0:
return
if command[0] == '#':
return
# Strip leading numbers from regression tests
if command[0].isdigit():
command = re.sub("^\d+", "", command).lstrip()
elements = command.split()
if not elements:
return
command_name = elements[0]; args = elements[1:]
if self.arg_error(command_name, len(args)):
return
if command_name in self.commands:
try:
self.commands[command_name](args)
except Exception as e:
self.debug_msg("Error executing command {}\n".format(str(e)))
self.debug_msg("Stack Trace:\n{}\n".format(traceback.format_exc()))
traceback.print_exc(file=sys.stdout)
raise e
else:
self.debug_msg("Unknown command: {}\n".format(command_name))
self.error('Unknown command')
sys.stdout.flush()
def arg_error(self, cmd, argnum):
"""
checker function for the number of arguments given to a command
Arguments
---------
cmd : str
the command name
argnum : int
number of parsed argument
Returns
-------
True if there was an argument error
False otherwise
"""
if cmd in self.argmap and self.argmap[cmd][0] != argnum:
self.error(self.argmap[cmd][1])
return True
return False
def debug_msg(self, msg = ''):
""" Write a msg to the debug stream """
if self._debug_mode:
sys.stderr.write(msg); sys.stderr.flush()
def error(self, error_msg = ''):
""" Send error msg to stdout and through the GTP connection. """
sys.stdout.write('? {}\n\n'.format(error_msg)); sys.stdout.flush()
def respond(self, response = ''):
""" Send msg to stdout """
sys.stdout.write('= {}\n\n'.format(response)); sys.stdout.flush()
def reset(self, size):
"""
Resets the state of the GTP to a starting board
Arguments
---------
size : int
the boardsize to reinitialize the state to
"""
self.board.reset(size)
self.go_engine.reset()
def protocol_version_cmd(self, args):
""" Return the GTP protocol version being used (always 2) """
self.respond('2')
def quit_cmd(self, args):
""" Quit game and exit the GTP interface """
self.respond()
exit()
def name_cmd(self, args):
""" Return the name of the player """
self.respond(self.go_engine.name)
def version_cmd(self, args):
""" Return the version of the player """
self.respond(self.go_engine.version)
def clear_board_cmd(self, args):
""" clear the board """
self.reset(self.board.size)
self.respond()
def boardsize_cmd(self, args):
"""
Reset the game and initialize with a new boardsize
Arguments
---------
args[0] : int
size of reinitialized board
"""
self.reset(int(args[0]))
self.respond()
def showboard_cmd(self, args):
self.respond('\n' + str(self.board.get_twoD_board()))
def showoptions_cmd(self,args):
options = dict()
options['komi'] = self.go_engine.komi
options['pattern'] = self.go_engine.pattern
options['selfatari'] = self.go_engine.selfatari
options['num_sim'] = self.go_engine.num_simulation
self.respond(options)
def komi_cmd(self, args):
"""
Set the komi for the game
Arguments
---------
args[0] : float
komi value
"""
self.go_engine.komi = float(args[0])
self.respond()
def known_command_cmd(self, args):
"""
Check if a command is known to the GTP interface
Arguments
---------
args[0] : str
the command name to check for
"""
if args[0] in self.commands:
self.respond("true")
else:
self.respond("false")
def list_commands_cmd(self, args):
""" list all supported GTP commands """
self.respond(' '.join(list(self.commands.keys())))
def set_free_handicap(self, args):
"""
clear the board and set free handicap for the game
Arguments
---------
args[0] : str
the move to handicap (e.g. B2)
"""
self.board.reset(self.board.size)
for point in args:
move = GoBoardUtil.move_to_coord(point, self.board.size)
point = self.board._coord_to_point(*move)
if not self.board.move(point, BLACK):
self.debug_msg("Illegal Move: {}\nBoard:\n{}\n".format(move, str(self.board.get_twoD_board())))
self.respond()
def legal_moves_cmd(self, args):
"""
list legal moves for current player
"""
color = self.board.current_player
legal_moves = GoBoardUtil.generate_legal_moves(self.board, color)
self.respond(GoBoardUtil.sorted_point_string(legal_moves, self.board.NS))
def num_sim_cmd(self, args):
self.go_engine.num_simulation = int(args[0])
self.respond()
def go_param_cmd(self, args):
valid_values = [0,1]
valid_params = ['selfatari','pattern']
param = args[0]
param_value = int(args[1])
if param not in valid_params:
self.error('Unkown parameters: {}'.format(param))
if param_value not in valid_values:
self.error('Argument 2 ({}) must be of type bool'.format(param_value))
if param ==valid_params[1]:
self.go_engine.pattern = param_value
elif param == valid_params[0]:
self.go_engine.selfatari = param_value
self.param_options[param] = param_value
self.respond()
def policy_moves_cmd(self, args):
"""
Return list of policy moves for the current_player of the board
"""
# Assignment4 - policy moves
# =========================================
move_prob_tuples = self.get_move_prob()
response = ' '.join([' '.join([t[0], '{:.5f}'.format(t[1])]) for t in move_prob_tuples])
self.respond(response)
# =========================================
def random_moves_cmd(self, args):
"""
Return list of random moves (legal, but not eye-filling)
"""
moves = GoBoardUtil.generate_random_moves(self.board)
if len(moves) == 0:
self.respond("Pass")
else:
self.respond(GoBoardUtil.sorted_point_string(moves, self.board.NS))
def feature_move_cmd(self, args):
move = None
if args[0]=='PASS':
move = 'PASS'
else:
move = GoBoardUtil.move_to_coord(args[0], self.board.size)
if move:
move = self.board._coord_to_point(move[0], move[1])
else:
self.error("Error in executing the move %s, check given move: %s"%(move,args[1]))
return
assert move != None
response = []
features = Feature.find_move_feature(self.board, move)
if features == None:
self.respond(response)
return
for f in features:
fn = Feature.find_feature_name(f)
if fn != None:
response.append(Feature.find_feature_name(f))
else:
response.append(self.board.neighborhood_33_pattern_shape(move))
r = '\n'.join(response)
self.respond(r)
def init_mm_file_header(self):
with open(self.mm_file_name ,'w') as header_writer:
header_writer.write('! 1080\n')
header_writer.write('8\n')
header_writer.write('2 Feature_Pass\n')
header_writer.write('1 Feature_Capture\n')
header_writer.write('2 Feature_Atari\n')
header_writer.write('1 Feature_SelfAtari\n')
header_writer.write('3 Feature_DistanceLine\n')
header_writer.write('8 Feature_DistancePrev\n')
header_writer.write('9 Feature_DistancePrevOwn\n')
header_writer.write('1054 Feature_Pattern\n')
header_writer.write('!\n')
header_writer.close()
self.init_mm_file = True
def feature_mm_cmd(self, args):
if self.init_mm_file == False:
self.init_mm_file_header()
assert self.init_mm_file == True
if len(self.board.moves) == 0:
with open('game_num.txt', 'a') as file:
self.num_game = self.num_game + 1
file.write('{}\n'.format(self.num_game))
file.close()
self.respond()
self.skip_counter = 0
return
# skip the first five moves in the game records, since they are randomly selected
if self.skip_counter <= 5:
self.skip_counter += 1
self.respond()
return
chosenMove = self.board.last_move
bd = self.board.copy()
if chosenMove != -1:
if chosenMove == None:
chosenMove = 'PASS'
bd.partial_undo_move()
Feature.write_mm_file(bd, chosenMove, self.mm_file_name)
if chosenMove == 'PASS':
bd.move(None, bd.current_player)
else:
bd.move(chosenMove, bd.current_player)
self.respond()
def gogui_analyze_cmd(self, args):
try:
self.respond("pstring/Legal Moves/legal_moves\n"
"pstring/Policy Moves/policy_moves\n"
"pstring/Random Moves/random_moves\n"
"none/Feature Move/feature_move %p\n"
"none/Features MM File/features_mm_file\n"
)
except Exception as e:
self.respond('Error: {}'.format(str(e)))
def play_cmd(self, args):
"""
play a move as the given color
Arguments
---------
args[0] : {'b','w'}
the color to play the move as
it gets converted to Black --> 1 White --> 2
color : {0,1}
board_color : {'b','w'}
args[1] : str
the move to play (e.g. A5)
"""
try:
board_color = args[0].lower()
board_move = args[1]
color= GoBoardUtil.color_to_int(board_color)
if args[1].lower()=='pass':
self.debug_msg("Player {} is passing\n".format(args[0]))
self.go_engine.update('pass')
#self.board.current_player = GoBoardUtil.opponent(color)
self.board.move(None, color)
self.respond()
return
move = GoBoardUtil.move_to_coord(args[1], self.board.size)
if move:
move = self.board._coord_to_point(move[0], move[1])
else:
self.error("Error in executing the move %s, check given move: %s"%(move,args[1]))
return
if not self.board.move(move, color):
self.respond("Illegal Move: {}".format(board_move))
return
else:
self.debug_msg("Move: {}\nBoard:\n{}\n".format(board_move, str(self.board.get_twoD_board())))
self.go_engine.update(move)
self.respond()
except Exception as e:
self.respond("illegal move: {} {} {}".format(board_color, board_move, str(e)))
def final_score_cmd(self, args):
self.respond(self.board.final_score(self.go_engine.komi))
def genmove_cmd(self, args):
"""
generate a move for the specified color
Arguments
---------
args[0] : {'b','w'}
the color to generate a move for
it gets converted to Black --> 1 White --> 2
color : {0,1}
board_color : {'b','w'}
"""
try:
board_color = args[0].lower()
color = GoBoardUtil.color_to_int(board_color)
self.debug_msg("Board:\n{}\nko: {}\n".format(str(self.board.get_twoD_board()),
self.board.ko_constraint))
# Assignment4 - policy probabilistic player
# =========================================
if color != self.board.current_player:
self.respond("Opponent's turn")
return
move_prob_tuples = self.get_move_prob()
if move_prob_tuples[0][0] == 'pass':
self.respond("pass")
self.go_engine.update('pass')
self.board.move(None, color)
return
# based on
# https://docs.python.org/3.5/library/random.html
population = [val for val, cnt in move_prob_tuples for _ in range(int(cnt*1e5))]
move = random.choice(population)
move = GoBoardUtil.move_to_coord(move, self.board.size)
move = self.board._coord_to_point(move[0], move[1])
# =========================================
if not self.board.check_legal(move, color):
move = self.board._point_to_coord(move)
board_move = GoBoardUtil.format_point(move)
self.respond("Illegal move: {}".format(board_move))
traceback.print_exc(file=sys.stdout)
self.respond('Error: {}'.format(str(e)))
raise RuntimeError("Illegal move given by engine")
# move is legal; play it
self.board.move(move, color)
self.debug_msg("Move: {}\nBoard: \n{}\n".format(move, str(self.board.get_twoD_board())))
move = self.board._point_to_coord(move)
board_move = GoBoardUtil.format_point(move)
self.respond(board_move)
except Exception as e:
self.respond('Error: {}'.format(str(e)))
# Assignment4
# =========================================
def get_move_prob(self):
moves = GoBoardUtil.generate_random_moves(self.board) # legal and not eye-filling
features = Feature.find_all_features(self.board)
gammas_sum = 0
move_gammas = dict()
if len(Features_weight) != 0:
for move in moves:
move_gammas[move] = Feature.compute_move_gamma(Features_weight, features[move])
gammas_sum += move_gammas[move]
# normalize to get probability
if gammas_sum:
for move in move_gammas.keys():
move_gammas[move] /= gammas_sum
if move_gammas and gammas_sum:
move_prob_tuples = [(self.board.point_to_string(k), v) for (k, v) in move_gammas.items()]
# sort list by probability and alphabetic order
# based on
# http://stackoverflow.com/questions/5212870/sorting-a-python-list-by-two-criteria
# answered by jaap on Stack Overflow http://stackoverflow.com/users/1186954/jaap
move_prob_tuples = sorted(move_prob_tuples, key=lambda k:(-k[1], k[0][0], k[0][1]))
else:
move_prob_tuples = list()
move_prob_tuples.append(('pass', 1))
return move_prob_tuples
#!/usr/bin/python3
from board import GoBoard
from board_util import GoBoardUtil, BLACK, WHITE, EMPTY, BORDER, FLOODFILL
import numpy as np
from Go5 import Go5Player
import time
board = GoBoard(4)
player = Go5Player(num_simulation=200, limit=100, exploration=np.sqrt(2))
player.MCTS.komi = 6.5
player.num_nodes = 5
cboard = board.copy()
print("\nrunning playout 200 times\n")
player.run(cboard, BLACK, print_info=True)
#time.sleep(30) # sleeping
player.num_simulation = 300
print("\nrunning it 300 more times\n")
cboard = board.copy()
player.run(cboard, BLACK, print_info=True)
#time.sleep(30)
print("\nrunning it 300 more times\n")
cboard = board.copy()
player.run(cboard, BLACK, print_info=True)
#time.sleep(30)
print("\nrunning it 300 more times\n")
cboard = board.copy()
player.run(cboard, BLACK, print_info=True)
class GtpConnection():
def __init__(self, go_engine, debug_mode=False):
"""
object that plays Go using GTP
Parameters
----------
go_engine: GoPlayer
a program that is capable of playing go by reading GTP commands
debug_mode: prints debug messages
"""
self.stdout = sys.stdout
sys.stdout = self
self._debug_mode = debug_mode
self.go_engine = go_engine
self.komi = 0
self.board = GoBoard(7)
# Assignment2 - 1.timelimit
self.timelimit = 1
signal.signal(signal.SIGALRM, self.timeout)
self.boarddic = {}
for d in range(0, 49):
self.boarddic[d] = {}
self.commands = {
"protocol_version": self.protocol_version_cmd,
"quit": self.quit_cmd,
"name": self.name_cmd,
"boardsize": self.boardsize_cmd,
"showboard": self.showboard_cmd,
"clear_board": self.clear_board_cmd,
"komi": self.komi_cmd,
"version": self.version_cmd,
"known_command": self.known_command_cmd,
"set_free_handicap": self.set_free_handicap,
"genmove": self.genmove_cmd,
"list_commands": self.list_commands_cmd,
"play": self.play_cmd,
"final_score": self.final_score_cmd,
"legal_moves": self.legal_moves_cmd,
# Assignment2 - 1.timelimit
"timelimit": self.timelimit_cmd,
# Assignment2 - 2.solve
"solve": self.solve_cmd,
}
# used for argument checking
# values: (required number or arguments, error message on argnum failure)
self.argmap = {
"boardsize": (1, 'Usage: boardsize INT'),
"komi": (1, 'Usage: komi FLOAT'),
"known_command": (1, 'Usage: known_command CMD_NAME'),
"set_free_handicap":
(1, 'Usage: set_free_handicap MOVE (e.g. A4)'),
"genmove": (1, 'Usage: genmove {w, b}'),
"play": (2, 'Usage: play {b, w} MOVE'),
"legal_moves": (1, 'Usage: legal_moves {w, b}'),
# Assignment2 - 1.timelimit
"timelimit": (1, "Usage: timelimit 1<=INT<=100"),
}
def __del__(self):
sys.stdout = self.stdout
def write(self, data):
self.stdout.write(data)
def flush(self):
self.stdout.flush()
def start_connection(self):
"""
start a GTP connection. This function is what continuously monitors
the user's input of commands.
"""
self.debug_msg("Start up successful...\n\n")
line = sys.stdin.readline()
while line:
self.get_cmd(line)
line = sys.stdin.readline()
def get_cmd(self, command):
"""
parse the command and execute it
Arguments
---------
command : str
the raw command to parse/execute
"""
if len(command.strip(' \r\t')) == 0:
return
if command[0] == '#':
return
# Strip leading numbers from regression tests
if command[0].isdigit():
command = re.sub("^\d+", "", command).lstrip()
elements = command.split()
if not elements:
return
command_name = elements[0]
args = elements[1:]
if command_name == "play" and self.argmap[command_name][0] != len(
args):
self.respond('illegal move: {} wrong number of arguments'.format(
args[0]))
return
if self.arg_error(command_name, len(args)):
return
if command_name in self.commands:
try:
self.commands[command_name](args)
except Exception as e:
self.debug_msg("Error executing command {}\n".format(str(e)))
self.debug_msg("Stack Trace:\n{}\n".format(
traceback.format_exc()))
raise e
else:
self.debug_msg("Unknown command: {}\n".format(command_name))
self.error('Unknown command')
sys.stdout.flush()
def arg_error(self, cmd, argnum):
"""
checker funciton for the number of arguments given to a command
Arguments
---------
cmd : str
the command name
argnum : int
number of parsed argument
Returns
-------
True if there was an argument error
False otherwise
"""
if cmd in self.argmap and self.argmap[cmd][0] > argnum:
self.error(self.argmap[cmd][1])
return True
return False
def debug_msg(self, msg=''):
""" Write a msg to the debug stream """
if self._debug_mode:
sys.stderr.write(msg)
sys.stderr.flush()
def error(self, error_msg=''):
""" Send error msg to stdout and through the GTP connection. """
sys.stdout.write('? {}\n\n'.format(error_msg))
sys.stdout.flush()
def respond(self, response=''):
""" Send msg to stdout """
sys.stdout.write('= {}\n\n'.format(response))
sys.stdout.flush()
def reset(self, size):
"""
Resets the state of the GTP to a starting board
Arguments
---------
size : int
the boardsize to reinitialize the state to
"""
if size != self.board.size:
self.boarddic = {}
for depth in range(0, size**2):
self.boarddic[depth] = {}
self.board.reset(size)
def protocol_version_cmd(self, args):
""" Return the GTP protocol version being used (always 2) """
self.respond('2')
def quit_cmd(self, args):
""" Quit game and exit the GTP interface """
self.respond()
exit()
def name_cmd(self, args):
""" Return the name of the player """
self.respond(self.go_engine.name)
def version_cmd(self, args):
""" Return the version of the player """
self.respond(self.go_engine.version)
def clear_board_cmd(self, args):
""" clear the board """
self.reset(self.board.size)
self.respond()
def boardsize_cmd(self, args):
"""
Reset the game and initialize with a new boardsize
Arguments
---------
args[0] : int
size of reinitialized board
"""
self.reset(int(args[0]))
self.respond()
def showboard_cmd(self, args):
self.respond('\n' + str(self.board.get_twoD_board()))
def komi_cmd(self, args):
"""
Set the komi for the game
Arguments
---------
args[0] : float
komi value
"""
self.komi = float(args[0])
self.respond()
def known_command_cmd(self, args):
"""
Check if a command is known to the GTP interface
Arguments
---------
args[0] : str
the command name to check for
"""
if args[0] in self.commands:
self.respond("true")
else:
self.respond("false")
def list_commands_cmd(self, args):
""" list all supported GTP commands """
self.respond(' '.join(list(self.commands.keys())))
def set_free_handicap(self, args):
"""
clear the board and set free handicap for the game
Arguments
---------
args[0] : str
the move to handicap (e.g. B2)
"""
self.board.reset(self.board.size)
for point in args:
move = GoBoardUtil.move_to_coord(point, self.board.size)
point = self.board._coord_to_point(*move)
if not self.board.move(point, BLACK):
self.debug_msg("Illegal Move: {}\nBoard:\n{}\n".format(
move, str(self.board.get_twoD_board())))
self.respond()
def legal_moves_cmd(self, args):
"""
list legal moves for the given color
Arguments
---------
args[0] : {'b','w'}
the color to play the move as
it gets converted to Black --> 1 White --> 2
color : {0,1}
board_color : {'b','w'}
"""
try:
board_color = args[0].lower()
color = GoBoardUtil.color_to_int(board_color)
moves = GoBoardUtil.generate_legal_moves(self.board, color)
self.respond(moves)
except Exception as e:
self.respond('Error: {}'.format(str(e)))
def play_cmd(self, args):
"""
play a move as the given color
Arguments
---------
args[0] : {'b','w'}
the color to play the move as
it gets converted to Black --> 1 White --> 2
color : {0,1}
board_color : {'b','w'}
args[1] : str
the move to play (e.g. A5)
"""
try:
board_color = args[0].lower()
board_move = args[1]
color = GoBoardUtil.color_to_int(board_color)
move = GoBoardUtil.move_to_coord(args[1], self.board.size)
if move:
move = self.board._coord_to_point(move[0], move[1])
else:
return
if not self.board.move(move, color):
return
self.respond()
except Exception as e:
self.respond("illegal move: {} {} {}".format(
board_color, board_move, str(e)))
def final_score_cmd(self, args):
self.respond(self.board.final_score(self.komi))
"""
Assignment2 - 3.genmove
"""
def genmove_cmd(self, args):
"""
try to generate a perfect move for the specified color
Arguments
---------
args[0] : {'b','w'}
the color to generate a move for
it gets converted to Black --> 1 White --> 2
color : {0,1}
board_color : {'b','w'}
"""
try:
board_color = args[0].lower()
color = GoBoardUtil.color_to_int(board_color)
_, move = self.solve()
if not move:
move = self.go_engine.get_move(self.board, color)
if move is None:
self.respond("resign")
return
if not self.board.check_legal(move, color):
move = self.board._point_to_coord(move)
board_move = GoBoardUtil.format_point(move)
self.respond("Illegal move: {}".format(board_move))
raise RuntimeError("Illegal move given by engine")
# move is legal; play it
self.board.move(move, color)
self.debug_msg("Move: {}\nBoard: \n{}\n".format(
move, str(self.board.get_twoD_board())))
move = self.board._point_to_coord(move)
board_move = GoBoardUtil.format_point(move)
self.respond(board_move)
except Exception as e:
self.respond('Error: {}'.format(str(e)))
# try:
# board_color = args[0].lower()
# color = GoBoardUtil.color_to_int(board_color)
# move = self.go_engine.get_move(self.board, color)
# if move is None:
# self.respond("pass")
# return
#
# if not self.board.check_legal(move, color):
# move = self.board._point_to_coord(move)
# board_move = GoBoardUtil.format_point(move)
# self.respond("Illegal move: {}".format(board_move))
# raise RuntimeError("Illegal move given by engine")
#
# # move is legal; play it
# self.board.move(move, color)
# self.debug_msg("Move: {}\nBoard: \n{}\n".format(move, str(self.board.get_twoD_board())))
# move = self.board._point_to_coord(move)
# board_move = GoBoardUtil.format_point(move)
# self.respond(board_move)
# except Exception as e:
# self.respond('Error: {}'.format(str(e)))
"""
Assignment2 - 1.timelimit
"""
def timelimit_cmd(self, args):
"""
set the maximum time to use for all following genmove or solve commands
the default value is 1
Arguments
---------
args[0] : int
timelimit
"""
try:
t = int(args[0])
if t in range(1, 101):
self.timelimit = t
self.respond()
else:
self.error(self.argmap["timelimit"][1])
except Exception as e:
self.respond("Error: {}".format(str(e)))
def timeout(self, signum, frame):
raise Exception("Timeout")
"""
Assignment2 - 2.solve
"""
def solve_cmd(self, args):
try:
winner, move = self.solve()
if move:
move = self.board._point_to_coord(move)
move = GoBoardUtil.format_point(move)
self.respond("{} {}".format(winner, move))
except Exception as e:
self.respond("Error: {}".format(str(e)))
"""
Helper function
"""
def solve(self):
board = self.board.copy()
depth = self.board.depth
try:
signal.alarm(self.timelimit)
#start = time.process_time()
result = self.negamaxBoolean(board, depth)
#self.respond("Time: {}".format(str(time.process_time() - start)))
signal.alarm(0)
except Exception as e:
if str(e) == "Timeout":
winner, move = "unknown", ""
else:
raise e
else:
if result[0]:
winner, move = GoBoardUtil.int_to_color(
self.board.to_play), result[1]
else:
winner, move = GoBoardUtil.int_to_color(
GoBoardUtil.opponent(self.board.to_play)), ""
return winner, move
# Based on https://webdocs.cs.ualberta.ca/~mmueller/courses/496-general/python/code/tic_tac_toe_solve.py
def negamaxBoolean(self, board, depth):
# optimize
board2D = board.get_twoD_board().tostring()
if board2D in self.boarddic[depth]:
if self.boarddic[depth][board2D][0] == board.to_play:
return True, self.boarddic[depth][board2D][1]
else:
return False, None
if board.get_winner():
return False, None
for m in GoBoardUtil.generate_legal_moves_fast(board, board.to_play):
backup = board.copy()
board.move(m, board.to_play)
success = not self.negamaxBoolean(board, depth + 1)[0]
board = backup
if success:
self.boardData(depth, board, m)
return True, m
return False, None
def boardData(self, depth, board, m):
# Store the {depth:{state: (winner, move), ...}, ...}
self.boarddic[depth][board.get_twoD_board().tostring()] = (
board.to_play, m)
# optimize 2
boardarr = board.get_twoD_board()
blist = []
blist.append(boardarr.tostring())
for i in range(0, 7):
emptyboard = GoBoard(board.size)
emptyboard.move(m, 1)
if i == 0:
b = np.rot90(boardarr)
if b.tostring() in blist:
continue
empty_move = np.rot90(emptyboard.get_twoD_board())
elif i == 1:
b = np.rot90(boardarr, 2)
if b.tostring() in blist:
continue
empty_move = np.rot90(emptyboard.get_twoD_board(), 2)
elif i == 2:
b = np.rot90(boardarr, 3)
if b.tostring() in blist:
continue
empty_move = np.rot90(emptyboard.get_twoD_board(), 3)
elif i == 3:
b = np.fliplr(boardarr)
if b.tostring() in blist:
continue
empty_move = np.fliplr(emptyboard.get_twoD_board())
elif i == 4:
b = np.flipud(boardarr)
if b.tostring() in blist:
continue
empty_move = np.flipud(emptyboard.get_twoD_board())
elif i == 5:
b = np.fliplr(np.rot90(boardarr))
if b.tostring() in blist:
continue
empty_move = np.fliplr(np.rot90(emptyboard.get_twoD_board()))
else:
b = np.flipud(np.rot90(boardarr))
if b.tostring() in blist:
continue
empty_move = np.flipud(np.rot90(emptyboard.get_twoD_board()))
blist.append(b.tostring())
index = np.where(empty_move == 1)
r, c = index[0][0] + 1, index[1][0] + 1
new_move = board._coord_to_point(r, c)
self.boarddic[depth][b.tostring()] = (board.to_play, new_move)