def custom_score_3(game: Board, player) -> float:
def future_open_move_locations(game, player):
future_step_locations = []
blank_spaces = game.get_blank_spaces()
player_location = game.get_player_location(player)
s_moves = []
s_moves.extend(second_moves_outside)
s_moves.extend(square_moves)
for move in s_moves:
location = tuple(map(sum, zip(player_location, move)))
if location in blank_spaces:
future_step_locations.append(location)
return future_step_locations
def future_open_move_locations_amount(game, player):
return len(future_open_move_locations(game, player))
if game.is_winner(player):
return INFINITY
if game.is_loser(player):
return NEG_INFINITY
future_step_locations_amount = future_open_move_locations_amount(game, player)
player_moves_count = len(game.get_legal_moves(player))
opponent_moves = len(game.get_legal_moves(game.get_opponent(player)))
return float(player_moves_count / 2 + future_step_locations_amount / 3 - opponent_moves / 2)
def min_value(self, board: Board, depth, alpha, beta):
"""
function MIN-VALUE(state, alpha, beta) returns a utility value
if TERMINAL-TEST(state) the return UTILITY(state)
v <- +infinity
for each a in ACTIONS(state) do
v <- MIN(v, MAX-VALUE(RESULT(state, a), alpha, beta))
if v <= alpha then return v
beta <- MIN(beta, v)
return v
"""
if self.time_left() < self.TIMER_THRESHOLD:
raise SearchTimeout()
mini = self
if board.is_winner(mini) or board.is_loser(mini):
return board.utility(mini)
if depth <= 0:
# what is the score function of the opponent? certainly not ours!
return self.score(board, mini)
min_value = math.inf
for move in board.get_legal_moves():
min_value = min(
min_value,
self.max_value(board.forecast_move(move), depth - 1, alpha,
beta))
if min_value <= alpha:
return min_value
beta = min(beta, min_value)
return min_value
def max_value(self, board: Board, depth, alpha, beta):
"""
function MAX-VALUE(state, alpha, beta) returns a utility value
if TERMINAL-TEST(state) the return UTILITY(state)
v <- -infinity
for each a in ACTIONS(state) do
v <- MAX(v, MIN-VALUE(RESULT(state, a), alpha, beta))
if v >= beta then return v
alpha <- MAX(alpha, v)
return v
"""
if self.time_left() < self.TIMER_THRESHOLD:
raise SearchTimeout()
maxi = self
if board.is_winner(maxi) or board.is_loser(maxi):
return board.utility(maxi)
if depth <= 0:
return self.score(board, maxi)
max_value = -math.inf
for move in board.get_legal_moves():
max_value = max(
max_value,
self.min_value(board.forecast_move(move), depth - 1, alpha,
beta))
if max_value >= beta:
return max_value
alpha = max(alpha, max_value)
return max_value
def custom_score(game: Board, player):
"""Calculate the heuristic value of a game state from the point of view
of the given player.
This should be the best heuristic function for your project submission.
Note: this function should be called from within a Player instance as
`self.score()` -- you should not need to call this function directly.
Parameters
----------
game : `isolation.Board`
An instance of `isolation.Board` encoding the current state of the
game (e.g., player locations and blocked cells).
player : object
A player instance in the current game (i.e., an object corresponding to
one of the player objects `game.__player_1__` or `game.__player_2__`.)
Returns
-------
float
The heuristic value of the current game state to the specified player.
"""
if game.is_loser(player):
return float("-inf")
if game.is_winner(player):
return float("inf")
opponent = game.get_opponent(player)
progress = blanks_left_percent(game)
if progress < 0.5:
if others_toe(game, player, opponent):
return 1
# close in on opponent
w, h = game.get_player_location(opponent)
y, x = game.get_player_location(player)
return float(-(h - y)**2 - (w - x)**2)
else:
# tread on his toes
multiplier = 1.0
if others_toe(game, player, opponent):
multiplier = 2.0
# if a toe is hit, we put some positive weight on players moves
own_moves = len(game.get_legal_moves(player))
opp_moves = len(game.get_legal_moves(opponent))
return float(own_moves - opp_moves) * multiplier
def min_value(self, board: Board, depth):
"""
function MIN-VALUE(state) returns a utility value
if TERMINAL-TEST(state) then return UTILITY(state)
v <- infinity
for each a in ACTIONS(state) do
v <- MIN(v, MAX-VALUE(RESULT(state, a)))
return v
"""
if self.time_left() < self.TIMER_THRESHOLD:
raise SearchTimeout()
mini = self
if board.is_winner(mini) or board.is_loser(mini):
return board.utility(mini)
if depth <= 0:
return self.score(board, mini)
min_value = math.inf
for move in board.get_legal_moves():
min_value = min(
min_value, self.max_value(board.forecast_move(move),
depth - 1))
return min_value
def custom_score(game: Board, player):
"""Calculate the heuristic value of a game state from the point of view
of the given player.
This should be the best heuristic function for your project submission.
Note: this function should be called from within a Player instance as
`self.score()` -- you should not need to call this function directly.
Parameters
----------
game : `isolation.Board`
An instance of `isolation.Board` encoding the current state of the
game (e.g., player locations and blocked cells).
player : object
A player instance in the current game (i.e., an object corresponding to
one of the player objects `game.__player_1__` or `game.__player_2__`.)
Returns
-------
float
The heuristic value of the current game state to the specified player.
results
=======
2 forecasts:
------------
Match # Opponent AB_Custom
Won | Lost
1 Random 10 | 0
2 MM_Open 7 | 3
3 MM_Center 8 | 2
4 MM_Improved 10 | 0
5 AB_Open 4 | 6
6 AB_Center 4 | 6
7 AB_Improved 4 | 6
--------------------------------------------------------------------------
Win Rate: 67.1%
1 forecast:
-----------
Match # Opponent AB_Custom
Won | Lost
1 Random 10 | 0
2 MM_Open 6 | 4
3 MM_Center 9 | 1
4 MM_Improved 8 | 2
5 AB_Open 4 | 6
6 AB_Center 7 | 3
7 AB_Improved 5 | 5
--------------------------------------------------------------------------
Win Rate: 70.0%
2 forecasts but only in potential endgame (last 15 rounds):
-----------------------------------------------------------
first try:
Match # Opponent AB_Custom
Won | Lost
1 Random 9 | 1
2 MM_Open 6 | 4
3 MM_Center 9 | 1
4 MM_Improved 6 | 4
5 AB_Open 5 | 5
6 AB_Center 3 | 7
7 AB_Improved 5 | 5
--------------------------------------------------------------------------
Win Rate: 61.4%
second try:
Match # Opponent AB_Custom
Won | Lost
1 Random 10 | 0
2 MM_Open 9 | 1
3 MM_Center 8 | 2
4 MM_Improved 7 | 3
5 AB_Open 6 | 4
6 AB_Center 6 | 4
7 AB_Improved 4 | 6
--------------------------------------------------------------------------
Win Rate: 71.4%
VS AB_Improved:
===============
2 forecasts but before potential endgame (except 15 rounds):
----------------------------------------------------------
Match # Opponent AB_Custom
Won | Lost
1 AB_Improved 22 | 18
--------------------------------------------------------------------------
Win Rate: 55.0%
"""
if game.is_loser(player):
return float("-inf")
if game.is_winner(player):
return float("inf")
opponent = game.get_opponent(player)
own_moves = []
opp_moves = []
if len(game.get_blank_spaces()) <= 25:
# 2 forecasts
for move_player in game.get_legal_moves(player):
moves_opponent = game.forecast_move(move_player).get_legal_moves(
opponent)
for move_opponent in moves_opponent:
ply = game.forecast_move(move_opponent)
own_moves += ply.get_legal_moves(player)
opp_moves += ply.get_legal_moves(opponent)
else:
own_moves = game.get_legal_moves(player)
opp_moves = game.get_legal_moves(opponent)
return float(len(own_moves) - len(opp_moves))