Pownie/backgammon
1
0
Fork 0
Code Issues 6 Pull Requests Packages Projects Releases Wiki Activity

Fixed potential bug in regards to scores in 2-ply calculation.

Browse Source
This commit is contained in:
Alexander Munch-Hansen 2018-04-25 00:51:04 +02:00
parent ea3f05846d
commit 8899c5c2d9
1 changed files with 30 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
network.py
View File

@ -113,6 +113,14 @@ class Network:
f.write(str(episode_count) + "\n")
def restore_model(self, sess):
"""
Restore a model for a session, such that a trained model and either be further trained or
used for evaluation
:param sess: Current session
:return: Nothing. It's a side-effect that a model gets restored for the network.
"""
if glob.glob(os.path.join(self.checkpoint_path, 'model.ckpt*.index')):
latest_checkpoint = tf.train.latest_checkpoint(self.checkpoint_path)
@ -150,6 +158,17 @@ class Network:
def make_move(self, sess, board, roll, player):
"""
Find the best move given a board, roll and a player, by finding all possible states one can go to
and then picking the best, by using the network to evaluate each state. The highest score is picked
for the 1-player and the max(1-score) is picked for the -1-player.
:param sess:
:param board: Current board
:param roll: Current roll
:param player: Current player
:return: A pair of the best state to go to, together with the score of that state
"""
legal_moves = Board.calculate_legal_states(board, player, roll)
moves_and_scores = [(move, self.eval_state(sess, self.board_trans_func(move, player))) for move in legal_moves]
scores = [x[1] if np.sign(player) > 0 else 1-x[1] for x in moves_and_scores]
@ -189,6 +208,10 @@ class Network:
# pythons reverse is in place and I can't call [:15] on it, without applying it to an object like so. Fuck.
best_fifteen = sorted(zero_ply_moves_and_scores, key=itemgetter(1))
# They're sorted from smallest to largest, therefore we wan't to reverse if the current player is 1, since
# player 1 wishes to maximize. It's not needed for player -1, since that player seeks to minimize.
if player == 1:
best_fifteen.reverse()
best_fifteen_boards = [x[0] for x in best_fifteen[:15]]
@ -198,12 +221,13 @@ class Network:
for a_board in best_fifteen_boards:
a_board_scores = []
for roll in all_rolls:
spec_roll_scores = []
all_rolls_boards = Board.calculate_legal_states(a_board, player*-1, roll)
spec_roll_scores.append(
[self.eval_state(sess, self.board_trans_func(new_board, player*-1)) for new_board in all_rolls_boards]
)
spec_roll_scores = [self.eval_state(sess, self.board_trans_func(new_board, player*-1))
for new_board in all_rolls_boards]
# We need 1-score for the -1 player
spec_roll_scores = [x if player == 1 else (1-x) for x in spec_roll_scores]
best_score = max(spec_roll_scores)
@ -372,8 +396,7 @@ class Network:
(random.randrange(1, 7), random.randrange(1, 7)),
player)
# print("The evaluation of the previous state:\n", self.eval_state(sess, self.board_trans_func(prev_board, player)))
# print("The evaluation of the current_state:\n", cur_board_value)
# adjust weights
sess.run(self.training_op,