igorcheb
/

REINFORCE-LunarLanderContinuous-v2

Reinforcement Learning

LunarLanderContinuous-v2

custom-implementation

Model card Files Files and versions Community

igorcheb commited on Jan 17, 2023

Commit

987b6f5

•

1 Parent(s): 3340525

Update agent_class.py

Files changed (1) hide show

agent_class.py +2 -2

agent_class.py CHANGED Viewed

@@ -2,8 +2,8 @@ import torch
 class ParameterisedPolicy(torch.nn.Module):
     """
-    REINFORCE RL agent class. Returns action when the ParameterisedPolicy.act(observation) method is used.
-    observation is a gym state vector.
     obs_len - length of the state vector.
     act_space_len - length of the action vector.
     """

 class ParameterisedPolicy(torch.nn.Module):
     """
+    REINFORCE RL agent class.
+    Use ParameterisedPolicy.act(observation) method to predict action based on input state.
     obs_len - length of the state vector.
     act_space_len - length of the action vector.
     """