update codes

codes/A2C/agent.py

@@ -10,12 +10,40 @@ Discription:
 Environment: 
 '''
 import torch.optim as optim
-from A2C.model import ActorCritic
+import torch.nn as nn
+import torch.nn.functional as F
+from torch.distributions import Categorical
+
+class ActorCritic(nn.Module):
+    ''' A2C网络模型，包含一个Actor和Critic
+    '''
+    def __init__(self, input_dim, output_dim, hidden_dim):
+        super(ActorCritic, self).__init__()
+        self.critic = nn.Sequential(
+            nn.Linear(input_dim, hidden_dim),
+            nn.ReLU(),
+            nn.Linear(hidden_dim, 1)
+        )
+        
+        self.actor = nn.Sequential(
+            nn.Linear(input_dim, hidden_dim),
+            nn.ReLU(),
+            nn.Linear(hidden_dim, output_dim),
+            nn.Softmax(dim=1),
+        )
+        
+    def forward(self, x):
+        value = self.critic(x)
+        probs = self.actor(x)
+        dist  = Categorical(probs)
+        return dist, value
 class A2C:
-    def __init__(self,n_states,n_actions,cfg) -> None:
+    ''' A2C算法
+    '''
+    def __init__(self,state_dim,action_dim,cfg) -> None:
         self.gamma = cfg.gamma
         self.device = cfg.device
-        self.model = ActorCritic(n_states, n_actions, cfg.hidden_size).to(self.device)
+        self.model = ActorCritic(state_dim, action_dim, cfg.hidden_size).to(self.device)
         self.optimizer = optim.Adam(self.model.parameters())
 
     def compute_returns(self,next_value, rewards, masks):