-
Notifications
You must be signed in to change notification settings - Fork 0
/
simulator.py
137 lines (94 loc) · 4.45 KB
/
simulator.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
# World
from world import World, PLAYER_1_NAME, PLAYER_2_NAME
# Arguement parser
import argparse
# Utilities
from utils import all_logging_disabled
# Logging
import logging
# Progress meters and progress bars
from tqdm import tqdm
# Numpy
import numpy as np
logging.basicConfig(format="%(levelname)s:%(message)s", level=logging.INFO)
logger = logging.getLogger(__name__)
# Get the arguements
def get_args():
parser = argparse.ArgumentParser()
parser.add_argument("--player_1", type=str, default="random_agent")
parser.add_argument("--player_2", type=str, default="random_agent")
parser.add_argument("--board_size", type=int, default=None)
parser.add_argument("--board_size_min",type=int,default=5,help="In autoplay mode, the minimum board size")
parser.add_argument("--board_size_max",type=int,default=11,help="In autoplay mode, the maximum board size")
parser.add_argument("--display", action="store_true", default=False)
parser.add_argument("--display_delay", type=float, default=0.4)
parser.add_argument("--display_save", action="store_true", default=False)
parser.add_argument("--display_save_path", type=str, default="plots/")
parser.add_argument("--autoplay", action="store_true", default=False)
parser.add_argument("--autoplay_runs", type=int, default=1000)
args = parser.parse_args()
return args
# Simulator
class Simulator:
# Initialization
def __init__(self, args):
self.args = args
# Reset
def reset(self, swap_players=False, board_size=None):
if board_size is None:
board_size = self.args.board_size
if swap_players:
player_1, player_2 = self.args.player_2, self.args.player_1
else:
player_1, player_2 = self.args.player_1, self.args.player_2
self.world = World(player_1=player_1,player_2=player_2,board_size=board_size,display_ui=self.args.display,display_delay=self.args.display_delay,display_save=self.args.display_save,display_save_path=self.args.display_save_path,autoplay=self.args.autoplay)
if self.world.initial_end:
logger.warning("Initialization failed! Reset the world again!")
self.reset()
# Run
def run(self, swap_players=False, board_size=None):
self.reset(swap_players=swap_players, board_size=board_size)
is_end, p0_score, p1_score = self.world.step()
while not is_end:
is_end, p0_score, p1_score = self.world.step()
logger.info(f"Run finished. Player {PLAYER_1_NAME}: {p0_score}, Player {PLAYER_2_NAME}: {p1_score}")
return p0_score, p1_score, self.world.p0_time, self.world.p1_time
# Autoplay with win %
def autoplay(self):
p1_win_count = 0
p2_win_count = 0
p1_times = []
p2_times = []
if self.args.display:
logger.warning("Since running autoplay mode, display will be disabled")
self.args.display = False
with all_logging_disabled():
for i in tqdm(range(self.args.autoplay_runs)):
swap_players = i % 2 == 0
board_size = np.random.randint(args.board_size_min, args.board_size_max)
p0_score, p1_score, p0_time, p1_time = self.run(swap_players=swap_players,board_size=board_size)
if swap_players:
p0_score, p1_score, p0_time, p1_time = (p1_score,p0_score,p1_time,p0_time)
if p0_score > p1_score:
p1_win_count += 1
elif p0_score < p1_score:
p2_win_count += 1
else: # Tie
p1_win_count += 1
p2_win_count += 1
p1_times.append(p0_time)
p2_times.append(p1_time)
logger.info(f"Player {PLAYER_1_NAME} win percentage: {p1_win_count / self.args.autoplay_runs} ({np.round(np.mean(p1_times), 5)} seconds/game)")
logger.info(f"Player {PLAYER_2_NAME} win percentage: {p2_win_count / self.args.autoplay_runs}, ({np.round(np.mean(p2_times), 5)} seconds/game)")
# Start of program
if __name__ == "__main__":
# Store command line arguements
args = get_args()
# Simulator object
simulator = Simulator(args)
# Auto play
if args.autoplay:
simulator.autoplay()
# Manual play
else:
simulator.run()