-
Notifications
You must be signed in to change notification settings - Fork 0
/
agent.py
40 lines (35 loc) · 1.37 KB
/
agent.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
import torch
import torch.nn as nn
import numpy as np
from torch.distributions.categorical import Categorical
def layer_init(layer, std=np.sqrt(2), bias_const=0.0):
torch.nn.init.orthogonal_(layer.weight, std)
torch.nn.init.constant_(layer.bias, bias_const)
return layer
class Agent(nn.Module):
def __init__(self, envs):
super(Agent, self).__init__()
# critic network: state -> action value
self.critic = nn.Sequential(
layer_init(nn.Linear(np.array(envs.single_observation_space.shape).prod(), 64)),
nn.Tanh(),
layer_init(nn.Linear(64, 64)),
nn.Tanh(),
layer_init(nn.Linear(64, 1), std=1.0),
)
# actor/policy network: state -> action logits
self.actor = nn.Sequential(
layer_init(nn.Linear(np.array(envs.single_observation_space.shape).prod(), 64)),
nn.Tanh(),
layer_init(nn.Linear(64, 64)),
nn.Tanh(),
layer_init(nn.Linear(64, envs.single_action_space.n), std=0.01),
)
def get_value(self, x):
return self.critic(x)
def get_action_and_value(self, x, action=None):
logits = self.actor(x)
probs = Categorical(logits=logits)
if action is None:
action = probs.sample()
return action, probs.log_prob(action), probs.entropy(), self.critic(x)