/
reinforce.py
99 lines (74 loc) · 2.73 KB
/
reinforce.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
import argparse
from itertools import count
import gym
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torch.autograd import Variable
from torch.distributions import Categorical
parser = argparse.ArgumentParser(description='PyTorch actor-critic example')
parser.add_argument('--gamma', type=float, default=0.99)
parser.add_argument('--seed', type=int, default=543, metavar='N')
parser.add_argument('--render', action='store_true')
args = parser.parse_args()
env = gym.make('CartPole-v0').unwrapped
env.seed(args.seed)
torch.manual_seed(args.seed)
class ActorCritic(nn.Module):
def __init__(self):
super().__init__()
self.affine1 = nn.Linear(4, 128)
self.action_head = nn.Linear(128, 2)
self.value_head = nn.Linear(128, 1)
def forward(self, x):
x = F.relu(self.affine1(x))
action_scores = self.action_head(x)
state_values = self.value_head(x)
return F.softmax(action_scores, dim=-1), state_values
def select_action(self, state, values, select_props):
state = torch.from_numpy(state).float()
props, value = self(Variable(state))
dist = Categorical(props)
action = dist.sample()
log_props = dist.log_prob(action)
values.append(value)
select_props.append(log_props)
return action.data[0]
model = ActorCritic()
optimizer = optim.Adam(model.parameters(), lr=3e-2)
def main():
for i_episode in count(1):
state = env.reset()
if args.render:
env.render()
values, select_props, policy_rewards = [], [], []
for t in range(10000):
action = model.select_action(state, values, select_props)
state, reward, done, _ = env.step(action)
policy_rewards.append(reward)
if done:
break
R, rewards = 0, []
for r in policy_rewards[::-1]:
R = r + args.gamma * R
rewards.insert(0, R)
rewards = np.asarray(rewards)
rewards = (rewards - rewards.mean()) / \
(rewards.std() + np.finfo(np.float32).eps)
value_loss, policy_loss = [], []
for value, prop, r in zip(values, select_props, rewards):
value_loss.append(F.smooth_l1_loss(
value, Variable(torch.Tensor([r]))))
reward = r - value.data[0]
policy_loss.append(-prop * reward)
loss = torch.cat(value_loss).sum() + torch.cat(policy_loss).sum()
optimizer.zero_grad()
loss.backward()
optimizer.step()
if i_episode % 10 == 0:
print('Episode {}\tLast length: {:5d}\t'.format(
i_episode, t))
if __name__ == '__main__':
main()