/
train1.py
133 lines (105 loc) · 4.07 KB
/
train1.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
from collections import deque
import numpy as np
from ple import PLE
from ple.games.waterworld import WaterWorld
from keras.layers import Dense
from keras.models import Sequential
import random
import matplotlib.pyplot as plt
class Agent:
def __init__(self, action_size):
self.epsilon_min = 0.01
self.epsilon_decay = 0.999999
self.create_model()
self.memory = deque(maxlen=2000)
self.gamma = 0.95 # discount rate
self.epsilon = 1 # exploration rate
self.learning_rate = 0.1
self.action_size = action_size
def create_model(self):
self.model = Sequential()
self.model.add(Dense(100, input_shape=(32,), activation='sigmoid'))
self.model.add(Dense(4, activation='sigmoid'))
self.model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
def remember(self, state, action, reward, next_state, done):
self.memory.append((state, action, reward, next_state, done))
def act(self, state):
if np.random.rand() <= self.epsilon:
return random.randrange(self.action_size)
act_values = self.model.predict(state)
return np.argmax(act_values[0])
def replay(self, batch_size):
minibatch = random.sample(self.memory, batch_size)
targets = list()
states = list()
for state, action, reward, next_state, done in minibatch:
target = reward
if not done:
target = reward + self.gamma * \
np.amax(self.model.predict(next_state)[0])
target_f = self.model.predict(state)
target_f[0][action] = target
targets.append(target_f[0])
states.append(state[0])
self.model.fit(np.array(states), np.array(targets), epochs=1, verbose=0)
if self.epsilon > self.epsilon_min:
self.epsilon *= self.epsilon_decay
rewards = {
"tick": -0.0, # each time the game steps forward in time the agent gets -0.1
"positive": 1, # each time the agent collects a green circle
"negative": -1.0, # each time the agent bumps into a red circle
}
# make a PLE instance.
# use lower fps so we can see whats happening a little easier
game = WaterWorld(width=256, height=256, num_creeps=15)
p = PLE(game, reward_values=rewards)
#
# p = PLE(game, fps=30, force_fps=False, display_screen=True,
# reward_values=rewards)
def process_state(current_state):
processed_state = list()
processed_state.append(current_state['player_x'])
processed_state.append(current_state['player_y'])
for creep in current_state['creep_pos']['GOOD']:
processed_state.append(-creep[0])
processed_state.append(-creep[1])
for creep in current_state['creep_pos']['BAD']:
processed_state.append(creep[0])
processed_state.append(creep[1])
return np.array((processed_state,))
p.init()
actions = p.getActionSet()[:-1]
agent = Agent(len(actions))
epochs = 10000000
game_duration = 1000
rewards = []
avg_rewards = []
epsilons = []
steps = []
step = 0
plt.ion()
for epoch in range(epochs):
p.reset_game()
for it in range(1000):
if p.game_over():
p.reset_game()
print "Score:" + str(p.score())
current_state = game.getGameState()
processed_current_state = process_state(current_state)
action = agent.act(processed_current_state)
reward = p.act(actions[action])
rewards.append(reward)
next_state = game.getGameState()
game_over = p.game_over()
processed_next_state = process_state(next_state)
agent.remember(processed_current_state, action, reward, processed_next_state, game_over)
if len(agent.memory) > 25:
agent.replay(25)
steps.append(epoch)
epsilons.append(agent.epsilon)
avg_rewards.append(np.average(rewards))
plt.plot(steps, avg_rewards, 'r')
plt.plot(steps, epsilons, 'g')
agent.model.save_weights("./models/model%d.h5" % epoch, overwrite=False)
print "Score: " + str(p.score())
print "Epsilon: " + str(agent.epsilon)