-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmulti_agent_handler.py
200 lines (147 loc) · 4.6 KB
/
multi_agent_handler.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
from bm_multi_env import *
import importlib
import numpy as np
#import lookahead_agent
#import random_agent
#import monte_carlo_agent
#import flee_agent
#import random
#import controlled_agent
import os
from time import sleep
import signal
import json
class TimeoutException(Exception): # Custom exception class
pass
def timeout_handler(signum, frame): # Custom signal handler
raise TimeoutException
#reading the configuration file
confi_json_file = "config.json"
with open ("config.json") as f:
config_data = json.load(f)
#dyanmically importing random
#module_name = "random_agent_class"
m1 = importlib.import_module(config_data["agent1"])
m2 = importlib.import_module(config_data["agent2"])
#having a solid_state
solid_state = {}
# Change the behavior of SIGALRM
signal.signal(signal.SIGALRM, timeout_handler)
# create the bomberman environment
env = Game(5,7)
#os.system('cls')
sleep(0.5)
# set play time
num_episodes = 3
max_turns = 200
timer_each_round = 3
# set agents in play
# available: lookahead_agent, random_agent, flee_agent, monte_carlo_agent
##agent1 = monte_carlo_agent
##agent2 = random_agent
#New way, by adding classes instead of method only
agent1 = m1.agent(1, env)
agent2 = m2.agent (2, env)
win_for_agent1 = 0
win_for_agent2 = 0
winner_json = {}
for i in range(num_episodes):
turn = 0
# initialize the map & players
board, players = env.reset()
#give the name of the player
env.players[0].name = agent1.name
env.players[1].name = agent2.name
# initialize variables
done = False
rewards = [0,0] # reward received per turn
total_rewards = [0,0] # cumulative rewards received
bomb_timer = env.MAX_TIMER
bomb_list = [] # a list of bomb objects in play and their properties
solid_state["players"] = players
# until game ends
while not done:
if turn > max_turns:
break
solid_state["board"] = board
solid_state["done"] = done
solid_state["bombs"] = bomb_list
solid_state["turn"] = turn
# render the game
env.render(True)
'''shhh! These are for future generation of the game.
print("board:")
print(board)
print("env:")
print(env)
print("validMoves:")
print(env.get_valid_actions(solid_state))
print("---")
'''
# This try/except loop ensures that
# you'll catch TimeoutException when it's sent.
signal.alarm(timer_each_round)
try:
# get player one's action
player_num1 = 1
p1_action = agent1.give_next_move(solid_state)
except TimeoutException:
#Give random move or no move continue # continue the for loop if function A takes more than 5 secon
p1_action = 0
print(">-<You snooze you lose agent1>-<")
else:
# Reset the alarm
signal.alarm(0)
#Start the timer. Once the time is over, a SIGALRM signal is sent.
signal.alarm(timer_each_round)
try:
# get player two's action
player_num2 = 2
p2_action = agent2.give_next_move(solid_state)
except TimeoutException:
p2_action = 0
print("<->You snooze you lose agent2<->")
#Give no move or random move continue # continue the for loop if function A takes more than 5 second
else:
# Reset the alarm
signal.alarm(0)
# perform action
actions = [p1_action, p2_action]
board, done, players, bomb_list = env.step(actions)
print(f"\n {agent1.name}: {p1_action}")
print(f"\n {agent2.name}: {p2_action}")
print(f"\n Turn: {turn}")
print(" --------------------")
print(f" Player 1 score: {players[0].score}")
print(" --------------------")
print(f" Player 2 score: {players[1].score}")
print(" --------------------")
turn +=1
sleep(0.2)
env.render(True)
env.convert_video() # it generates the video
if players[0].score > players[1].score:
print(f" Game over. {agent1.name} wins round{i+1}.")
win_for_agent1 = win_for_agent1 + 1
elif players[0].score < players[1].score:
print(f" Game over. {agent2.name} wins round{i+1}.")
win_for_agent2 = win_for_agent2 + 1
else:
print("Game over. It's a tie.")
#sleep(3)
#break
sleep(3)
if (i + 1) == num_episodes:
#this means it is in the last round
if (win_for_agent1 > win_for_agent2):
print(f" The winner is: {agent1.name}")
winner_json["winner"] = config_data["agent1"]
elif (win_for_agent2 > win_for_agent1):
print(f" The winner is: {agent2.name}")
winner_json["winner"] = config_data["agent2"]
else:
print("Agents need to play one more time!")
i = i - 1
sleep(0.2)
with open('winner.json', 'w') as outfile:
json.dump(winner_json, outfile)