gym-examples 3.0.79__py3-none-any.whl → 3.0.80__py3-none-any.whl

gym_examples/__init__.py

@@ -6,4 +6,4 @@ register(
      max_episode_steps=50,
 )
 
-__version__ = "3.0.79"
+__version__ = "3.0.80"

gym_examples/envs/wsn_env.py

@@ -40,6 +40,9 @@ class Attention(nn.Module):
         x = self.linear2(x)  # Pass the result through another linear layer
         return x
 
+net = Attention(input_dim, output_dim)
+net = net.double()  # Convert the weights to Double
+
 class WSNRoutingEnv(gym.Env):
     def __init__(self, n_sensors = 20, coverage_radius=(upper_bound - lower_bound)/4):
 
@@ -121,8 +124,8 @@ class WSNRoutingEnv(gym.Env):
                 self.total_latency += self.packet_latency[i] + latency_per_hop
                 self.packet_latency[i] = 0
 
-                # rewards[i] = self.compute_individual_rewards(i, action)
-                rewards[i] = np.ones(input_dim)  # Reward for transmitting data to the base station
+                rewards[i] = self.compute_individual_rewards(i, action)
+                # rewards[i] = np.ones(input_dim)  # Reward for transmitting data to the base station
                 dones[i] = True       
             else:
                 distance = np.linalg.norm(self.sensor_positions[i] - self.sensor_positions[action])
@@ -149,8 +152,8 @@ class WSNRoutingEnv(gym.Env):
                 self.number_of_packets[action] += self.number_of_packets[i]
             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] = np.mean(rewards[i])
+            rewards[i] = self.compute_attention_rewards(rewards[i])
+            # rewards[i] = np.mean(rewards[i])
         for i in range(self.n_sensors):
             if (self.remaining_energy[i] <= 0) or (self.number_of_packets[i] <= 0):
                 dones[i] = True
@@ -371,10 +374,7 @@ class WSNRoutingEnv(gym.Env):
         '''
         Compute the attention-based rewards
         '''
-        input_dim = len(rewards)
         rewards = torch.tensor(rewards, dtype=torch.double)
-        net = Attention(input_dim, output_dim)
-        net = net.double()  # Convert the weights to Double
         final_reward = net(rewards)
         return final_reward 
 

{gym_examples-3.0.79.dist-info → gym_examples-3.0.80.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gym-examples
-Version: 3.0.79
+Version: 3.0.80
 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-3.0.80.dist-info/RECORD

@@ -0,0 +1,7 @@
+gym_examples/__init__.py,sha256=hNQ05dR9XWqWYLWTRWQYrb94dzR8CeLomp79s3ty4pc,193
+gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
+gym_examples/envs/wsn_env.py,sha256=0-eIZjWC5qOwEcK3wJsdP9McFn4gptdQdoiI-I5QOlc,20386
+gym_examples-3.0.80.dist-info/METADATA,sha256=FgRrwMYEQGezDhVqtqUdE9HVpABQZtHSd9f3vXMEFRU,411
+gym_examples-3.0.80.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+gym_examples-3.0.80.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
+gym_examples-3.0.80.dist-info/RECORD,,

gym_examples-3.0.79.dist-info/RECORD

@@ -1,7 +0,0 @@
-gym_examples/__init__.py,sha256=Zx6apdel9zNMkb769ArqkX8c7CG_d8CHIpz4BQoJUnc,193
-gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
-gym_examples/envs/wsn_env.py,sha256=z4l1ctB9RlQmtxwb1Nl-QGtERQ1C_YVVhacx8fzfp7o,20434
-gym_examples-3.0.79.dist-info/METADATA,sha256=Tk9FJq9C_mAWmJoehqotrD_lB16I1t70kybNMWtHaVA,411
-gym_examples-3.0.79.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-gym_examples-3.0.79.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
-gym_examples-3.0.79.dist-info/RECORD,,