-
Notifications
You must be signed in to change notification settings - Fork 1
/
plotGrpahs.py
120 lines (95 loc) · 2.86 KB
/
plotGrpahs.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
import matplotlib.pyplot as plt
import numpy as np
rewards_hardcoded = np.load("episodeReward_2.npy")
rewards_random = np.load("episodeReward_1.npy")
rewards_dqn = np.load("data/episodeReward_DQN0.npy")
rewards_doubleduelingdqn = np.load("data/episodeReward_DoubleDuelingDQN0.npy")
rewards_doubledqn = np.load("data/episodeReward_DoubleDQN0.npy")
rewards_duelingdqn = np.load("data/episodeReward_DuelingDQN0.npy")
reward_dqn_2agents = np.load("data/episodeReward_DQN2agents0.npy")
reward_dddqn_2agents = np.load("data/episodeReward_DDDQN2agents0.npy")
def get_mean(rewards):
r = [0]*100
for i in range(100,rewards.shape[0]):
r.append(np.mean(np.array(rewards[i-100:i])))
return np.array(r)
rewards_hardcoded_mean = get_mean(rewards_hardcoded)
rewards_random_mean = get_mean(rewards_random)
rewards_dqn_mean = get_mean(rewards_dqn)
rewards_doubleduelingdqn_mean = get_mean(rewards_doubleduelingdqn)
rewards_doubledqn_mean = get_mean(rewards_doubledqn)
rewards_duelingdqn_mean = get_mean(rewards_duelingdqn)
reward_dqn_2agents_mean = get_mean(reward_dqn_2agents)
reward_dddqn_2agents_mean = get_mean(reward_dddqn_2agents)
plt.figure(1)
plt.clf()
plt.title('Score(Hard Coded Path)')
plt.xlabel('Episode')
plt.ylabel('Reward')
plt.plot(rewards_hardcoded)
plt.plot(rewards_hardcoded_mean)
plt.legend(['Episode reward','100 Episode avg'])
plt.show()
plt.figure(2)
plt.clf()
plt.title('Score(Random)')
plt.xlabel('Episode')
plt.ylabel('Reward')
plt.plot(rewards_random)
plt.plot(rewards_random_mean)
plt.legend(['Episode reward','100 Episode avg'])
plt.show()
plt.figure(3)
plt.clf()
plt.title('Score(DQN)')
plt.xlabel('Episode')
plt.ylabel('Reward')
plt.plot(rewards_dqn)
plt.plot(rewards_dqn_mean)
plt.legend(['Episode reward','100 Episode avg'])
plt.show()
plt.figure(4)
plt.clf()
plt.title('Score(DoubleDQN)')
plt.xlabel('Episode')
plt.ylabel('Reward')
plt.plot(rewards_doubledqn)
plt.plot(rewards_doubledqn_mean)
plt.legend(['Episode reward','100 Episode avg'])
plt.show()
plt.figure(5)
plt.clf()
plt.title('Score(DuelingDQN)')
plt.xlabel('Episode')
plt.ylabel('Reward')
plt.plot(rewards_duelingdqn)
plt.plot(rewards_duelingdqn_mean)
plt.legend(['Episode reward','100 Episode avg'])
plt.show()
plt.figure(6)
plt.clf()
plt.title('Score(DoubleDuelingDQN)')
plt.xlabel('Episode')
plt.ylabel('Reward')
plt.plot(rewards_doubleduelingdqn)
plt.plot(rewards_doubleduelingdqn_mean)
plt.legend(['Episode reward','100 Episode avg'])
plt.show()
plt.figure(7)
plt.clf()
plt.title('Score(DQN_2agents)')
plt.xlabel('Episode')
plt.ylabel('Reward')
plt.plot(reward_dqn_2agents)
plt.plot(reward_dqn_2agents_mean)
plt.legend(['Episode reward','100 Episode avg'])
plt.show()
plt.figure(8)
plt.clf()
plt.title('Score(DDDQN_2agents)')
plt.xlabel('Episode')
plt.ylabel('Reward')
plt.plot(reward_dddqn_2agents)
plt.plot(reward_dddqn_2agents_mean)
plt.legend(['Episode reward','100 Episode avg'])
plt.show()