Update Agent_class.py
Browse files- Agent_class.py +4 -4
Agent_class.py
CHANGED
@@ -2,7 +2,7 @@ import torch
|
|
2 |
|
3 |
class ParameterisedPolicy(torch.nn.Module):
|
4 |
"""
|
5 |
-
REINFORCE RL agent class. Returns action when the ParameterisedPolicy.act(observation) is used.
|
6 |
observation is a gym state vector.
|
7 |
obs_len - length of the state vector
|
8 |
act_space_len - length of the action vector
|
@@ -10,8 +10,6 @@ class ParameterisedPolicy(torch.nn.Module):
|
|
10 |
"""
|
11 |
def __init__(self, obs_len=8, act_space_len=2):
|
12 |
super().__init__()
|
13 |
-
self.deterministic = False
|
14 |
-
self.continuous = True
|
15 |
self.obs_len = obs_len
|
16 |
self.act_space_len = act_space_len
|
17 |
self.lin_1 = torch.nn.Linear(self.obs_len, 256)
|
@@ -41,7 +39,9 @@ class ParameterisedPolicy(torch.nn.Module):
|
|
41 |
return mu, sigma
|
42 |
|
43 |
def act(self, observation):
|
44 |
-
|
|
|
|
|
45 |
(mus, sigmas) = self.forward(torch.tensor(observation, dtype=torch.float32))
|
46 |
m = torch.distributions.normal.Normal(mus, sigmas)
|
47 |
action = m.sample().detach().numpy()
|
|
|
2 |
|
3 |
class ParameterisedPolicy(torch.nn.Module):
|
4 |
"""
|
5 |
+
REINFORCE RL agent class. Returns action when the ParameterisedPolicy.act(observation) method is used.
|
6 |
observation is a gym state vector.
|
7 |
obs_len - length of the state vector
|
8 |
act_space_len - length of the action vector
|
|
|
10 |
"""
|
11 |
def __init__(self, obs_len=8, act_space_len=2):
|
12 |
super().__init__()
|
|
|
|
|
13 |
self.obs_len = obs_len
|
14 |
self.act_space_len = act_space_len
|
15 |
self.lin_1 = torch.nn.Linear(self.obs_len, 256)
|
|
|
39 |
return mu, sigma
|
40 |
|
41 |
def act(self, observation):
|
42 |
+
"""
|
43 |
+
Method returns action when gym state vector is passed.
|
44 |
+
"""
|
45 |
(mus, sigmas) = self.forward(torch.tensor(observation, dtype=torch.float32))
|
46 |
m = torch.distributions.normal.Normal(mus, sigmas)
|
47 |
action = m.sample().detach().numpy()
|