-
Notifications
You must be signed in to change notification settings - Fork 1
/
rotabot.py
254 lines (228 loc) · 9.89 KB
/
rotabot.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
# -*- coding: utf-8 -*-
"""
Created on Wed Apr 8 13:20:33 2015
@author: killerdigby
"""
import requests
import random
from itertools import permutations as perm
import rota
class RotaBot():
def __init__(self):
self._circle = [0,1,2,5,8,7,6,3,0,1]
self._cross = [(0,4,8),(1,4,7),(2,4,6),(3,4,5)]
self._scores = {board:0.0 for board in perm('pppccc---',9)}
self._player_boards = {}
self._comp_boards = {}
self.initialize_scores()
self.inititialize_comp_boards()
def initialize_scores(self):
"""
Assign desirability value to boards for move selection
"""
for board in self._scores:
winner = self.check_winner(board)
if winner == 'c':
self._scores[tuple(board)] -= 3.0
moves = self.available_moves(board,'c')
for move in moves:
new_board = self.change_board(board,move,'c')
self._scores[tuple(new_board)] -= 2.0
inner_moves = self.available_moves(new_board,'c')
if winner == 'p':
self._scores[tuple(board)] += 3.0
moves = self.available_moves(board,'p')
for move in moves:
new_board = self.change_board(board,move,'p')
self._scores[tuple(new_board)] += 2.0
inner_moves = self.available_moves(new_board,'p')
for inner_move in inner_moves:
inner_board = self.change_board(new_board,inner_move,'p')
self._scores[tuple(inner_board)] += 1.0
def inititialize_comp_boards(self):
for board in self._scores:
if self._scores[board] <= 0:
player_pos = []
for idx in range(9):
if board[idx] == 'p':
player_pos.append(idx)
self._comp_boards[tuple(player_pos)] = self._scores[board]
def check_winner(self, board):
"""
Returns the winner
"""
for pos in range(8):
if board[self._circle[pos]] != '-':
if board[self._circle[pos]] == board[self._circle[pos+1]] ==board[self._circle[pos+2]]:
return board[self._circle[pos]]
for pos in self._cross:
if board[pos[0]] != '-':
if board[pos[0]] == board[pos[1]] == board[pos[2]]:
return board[pos[0]]
def change_board(self, board, move, player):
"""
Move a piece
"""
fromm, to = move
board = list(board)
board[fromm] = '-'
board[to] = player
return ''.join(board)
def available_moves(self, board, player):
moves = [0,1,2,5,8,7,6,3]
current = [idx for idx in range(9) if board[idx] == player]
#create list of (from, to moves
available = []
for pos in current:
if pos != 4:
idx = moves.index(pos)
if board[moves[idx-1]] == '-':
available.append((moves[idx],moves[idx-1]))
if board[moves[(idx+1)%8]] == '-':
available.append((moves[idx], moves[(idx+1)%8]))
if board[4] == '-':
available.append((moves[idx], 4))
elif pos == 4:
for idx in range(9):
if board[idx] == '-':
available.append((pos, idx))
return available
def update_scores(self, scores, board):
"""
Scores board and updates scores
"""
winner = self.check_winner(board)
increment = 1.0
if winner:
if winner == 'c':
increment = 1.2
other = 'p'
else:
other = 'c'
for idx in range(9):
if board[idx] == winner:
scores[idx] += increment * 1.0
elif board[idx] == other:
scores[idx] -= 1.0
def get_best_place(self, board, scores):
"""
for place score
"""
current_best = (None,0)
for idx in range(9):
if board[idx] == '-':
if current_best[0] == None:
current_best = (idx, scores[idx])
elif scores[idx] > current_best[1]:
current_best = (idx, scores[idx])
return current_best[0]
def mc_placing_trial(self, board):
"""
Randomly assign remaining player and computer choices
"""
board = list(board)
empty_spaces = [idx for idx in range(9) if board[idx] == u'-']
while board.count('p') < 3:
move = empty_spaces.pop(random.randrange(len(empty_spaces)))
board[move] = 'p'
if board.count('c') < 3:
move = empty_spaces.pop(random.randrange(len(empty_spaces)))
board[move] = 'c'
return "".join(board)
def mc_place(self, board):
"""
Returns best board position to place player piece
"""
#score of each position initiialized to 0
scores = [0.0] * 9
if board.count('-') == 9 :
ans = random.choice(self._circle)
print ans
return ans
elif board.count('c') == 3:
if board.count('p') == 2:
empty = [idx for idx in range(9) if board[idx] == '-']
available = []
for idx in empty:
new_board = list(board)
new_board[idx] = 'p'
available.append(("".join(new_board), idx))
best_moves = []
for move in available:
if best_moves == []:
best_moves.append((self._scores[tuple(move[0])], move[1]))
elif self._scores[tuple(move[0])] == best_moves[0][0]:
best_moves.append((self._scores[tuple(move[0])], move[1]))
elif self._scores[tuple(move[0])] > best_moves[0][0]:
best_moves = [(self._scores[tuple(move[0])], move[1])]
best_move = random.choice(best_moves)
return best_move[1]
for dummy_trial in range(10000):
played_board = self.mc_placing_trial(board)
self.update_scores(scores, played_board)
return self.get_best_place(board, scores)
def learned_player_move(self, board):
available = self.available_moves(board, 'p')
best_moves=[]
for move in available:
result_board = self.change_board(board, move,'p')
if best_moves == []:
best_moves.append((self._scores[tuple(result_board)], move))
elif self._scores[tuple(result_board)] == best_moves[0][0]:
best_moves.append((self._scores[tuple(result_board)], move))
elif self._scores[tuple(result_board)] > best_moves[0][0]:
best_moves = [(self._scores[tuple(result_board)], move)]
best_move = random.choice(best_moves)
return best_move[1]
def play(self):
"""
hear of the beast...don't look it's nasty
"""
found_hash=''
trials = 0
while found_hash == '':
try:
game = rota.RotaAPI()
res = game.json
board = game.json['data']['board']
trials +=1
while res['status'] == 'success':
print res
move=self.mc_place(board) + 1
res = game.place(move)
print res
if res['status'] == 'success':
board = res['data']['board']
count = 0
if board.count('p') == 3:
board_history = [board]
while res['status'] == 'success' and found_hash == '':
old_board = res['data']['board']
board_history.append(old_board)
fromm, to = self.learned_player_move(old_board)
res = game.move(fromm + 1, to + 1)
print res
try:
if res['data']['computer_wins'] == 1:
self._scores[tuple(res['data']['board'])] -= 3
amt = 2
for board in board_history[::-1]:
if board:
self._scores[tuple(board)] -= amt
amt -= 1
print count
break
except KeyError:
break
count+=1
if len(board_history) > 2:
board_history.pop(0)
try:
found_hash = res['data']['hash']
print found_hash
except KeyError:
pass
except requests.exceptions.ConnectionError:
pass
x= RotaBot()
x.play()