gym-examples 3.0.757__py3-none-any.whl → 3.0.759__py3-none-any.whl

gym_examples/__init__.py

@@ -5,4 +5,4 @@ register(
      entry_point="gym_examples.envs:WSNRoutingEnv",
 )
 
-__version__ = "3.0.757"
+__version__ = "3.0.759"

gym_examples/envs/wsn_env.py

@@ -51,8 +51,8 @@ class ScalarAttentionModel(nn.Module):
         scalar_output = self.output_layer(attention_output)
         return scalar_output
 
-# net = ScalarAttentionModel(input_dim)
-# net = net.double()  # Convert the weights to Double
+net = ScalarAttentionModel(input_dim)
+net = net.double()  # Convert the weights to Double
 
 class WSNRoutingEnv(gym.Env):
 
@@ -212,7 +212,7 @@ class WSNRoutingEnv(gym.Env):
                 
             self.number_of_packets[i] = 0 # Reset the number of packets of the sensor i
             # Calculate final reward
-            # rewards[i] = self.compute_attention_rewards(rewards[i])
+            rewards[i] = self.compute_attention_rewards(rewards[i])
             # rewards[i] = np.mean(rewards[i])
         
         # Integrate the mobility of the sensors
@@ -226,8 +226,7 @@ class WSNRoutingEnv(gym.Env):
         self.get_metrics()
 
         # rewards = [reward.item() if isinstance(reward, torch.Tensor) else reward for reward in rewards] # Convert the reward to a float
-        rewards = self.compute_attention_rewards(rewards, self.n_sensors)
-        # rewards = np.mean(rewards)
+        rewards = np.mean(rewards)
 
         for i in range(self.n_sensors):
             if not dones[i]:
@@ -464,20 +463,21 @@ class WSNRoutingEnv(gym.Env):
         return np.clip(normalized_throughput, 0, 1)
     
 
-    def compute_attention_rewards(self, rewards, n_sensors):
+    def compute_attention_rewards(self, reward):
         '''
         Compute the attention-based rewards
         '''
-        net = ScalarAttentionModel(n_sensors)
-        net = net.double()  # Convert the weights to Double
-        final_reward = []
-        for i in range(len(rewards[0])):
-            rewards_i = [reward[i] for reward in rewards]
-            rewards_i = torch.tensor(rewards_i, dtype=torch.double)
-            rewards_i = rewards_i.unsqueeze(0) # Add batch dimension
-            final_reward.append(net(rewards_i).item())
-        
-        return np.mean(final_reward) 
+        rewards_i = torch.tensor(reward, dtype=torch.double)
+        rewards_i = rewards_i.unsqueeze(0) # Add batch dimension
+        print(f"rewards_i: {rewards_i}")
+        print(f"rewards_i shape: {rewards_i.shape}")
+        print(f"rewards_i type: {rewards_i.dtype}")
+        print(f"net(rewards_i): {net(rewards_i)}")
+        print(f"net(rewards_i) shape: {net(rewards_i).shape}")
+        print(f"net(rewards_i) type: {net(rewards_i).dtype}")
+        print(f"net(rewards_i).item(): {net(rewards_i).item()}")
+        raise Exception("Stop here")
+        return np.mean(net(rewards_i).item()) 
 
 
     # def compute_attention_reward(self, rewards):

{gym_examples-3.0.757.dist-info → gym_examples-3.0.759.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gym-examples
-Version: 3.0.757
+Version: 3.0.759
 Summary: A custom environment for multi-agent reinforcement learning focused on WSN routing.
 Home-page: https://github.com/gedji/CODES.git
 Author: Georges Djimefo

gym_examples-3.0.759.dist-info/RECORD

@@ -0,0 +1,7 @@
+gym_examples/__init__.py,sha256=ZTeyp0SinvYP9QhqAJgRCJyDVdSqynNlieriu3lweGA,166
+gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
+gym_examples/envs/wsn_env.py,sha256=Z0W7fcJ696BlqAUnRNoZi6Qg9tJcvBhLvv_UHGCJYOs,26980
+gym_examples-3.0.759.dist-info/METADATA,sha256=zC4tuoN4KrjVHrfvi5ZCP5EYKmmv5CcXRvH_jQAZCfc,412
+gym_examples-3.0.759.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+gym_examples-3.0.759.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
+gym_examples-3.0.759.dist-info/RECORD,,

gym_examples-3.0.757.dist-info/RECORD

@@ -1,7 +0,0 @@
-gym_examples/__init__.py,sha256=wTbccbijA4S2bCClYZUAIV2wk4v8I02ORM5B8uPEouc,166
-gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
-gym_examples/envs/wsn_env.py,sha256=09K0IztE2cXxl9PChKgDaLVqPF2KNQ-Pav_h0PDdYNo,26947
-gym_examples-3.0.757.dist-info/METADATA,sha256=-WQV4YWJ9pbVuWL4z_MnISbNBEm7pD-4jZjF_S-SNuA,412
-gym_examples-3.0.757.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-gym_examples-3.0.757.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
-gym_examples-3.0.757.dist-info/RECORD,,