gym-examples 2.0.30__py3-none-any.whl → 2.0.32__py3-none-any.whl

gym_examples/__init__.py

@@ -6,4 +6,4 @@ register(
      max_episode_steps=50,
 )
 
-__version__ = "2.0.30"
+__version__ = "2.0.32"

gym_examples/envs/wsn_env.py

@@ -115,9 +115,10 @@ class WSNRoutingEnv(gym.Env):
             # Calculate final reward
             # rewards_individual = torch.tensor(rewards[i], dtype=torch.double)                
             # final_reward = net(rewards_individual)
-            final_reward = np.sum(rewards[i])
+            final_reward = np.sum(rewards[i]) 
             rewards[i] = final_reward
-
+        rewards *= self.network_reward_consumption_energy() * self.network_reward_dispersion_remaining_energy()
+        
         for i in range(self.n_sensors):
             if (self.remaining_energy[i] <= 0) or (self.number_of_packets[i] <= 0):
                 dones[i] = True
@@ -234,8 +235,11 @@ class WSNRoutingEnv(gym.Env):
         '''
         Compute the reward based on the number of packets of the receiver
         '''
-        max_number_of_packets = self.n_sensors * initial_number_of_packets 
-        normalized_number_of_packets = self.number_of_packets[action] / max_number_of_packets
+        max_number_of_packets = self.n_sensors * initial_number_of_packets
+        if action == self.n_sensors:
+            normalized_number_of_packets = 0
+        else: 
+            normalized_number_of_packets = self.number_of_packets[action] / max_number_of_packets
 
         return 1 - normalized_number_of_packets
 
@@ -251,6 +255,29 @@ class WSNRoutingEnv(gym.Env):
 
         return [reward_angle, reward_distance, reward_consumption_energy, reward_dispersion_remaining_energy, reward_number_of_packets]
 
+    def network_reward_dispersion_remaining_energy(self):
+        '''
+        Compute the reward based on the standard deviation of the remaining energy at the network level
+        '''
+        dispersion_remaining_energy = np.std(self.remaining_energy)
+        # Normalize the standard deviation of the remaining energy
+        max_dispersion_remaining_energy = initial_energy / 2 # maximum standard deviation of the remaining energy if n_sensors is even
+        normalized_dispersion_remaining_energy = dispersion_remaining_energy / max_dispersion_remaining_energy
+
+        return np.clip(1 - normalized_dispersion_remaining_energy, 0, 1)
+    
+    def network_reward_consumption_energy(self):
+        '''
+        Compute the reward based on the total energy consumption (transmission, reception) at the network level
+        '''
+        total_energy = np.sum(self.consumption_energy)
+        # Normalize the total energy consumption
+        max_transmission_energy = self.transmission_energy(self.n_sensors * initial_number_of_packets, self.coverage_radius)
+        max_reception_energy = self.reception_energy(self.n_sensors * initial_number_of_packets)
+        normalized_total_energy = total_energy / (max_transmission_energy + max_reception_energy)
+
+        return 1 - normalized_total_energy
+
     def integrate_mobility(self):
         '''
         Integrate the mobility of the sensors after each step

{gym_examples-2.0.30.dist-info → gym_examples-2.0.32.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gym-examples
-Version: 2.0.30
+Version: 2.0.32
 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-2.0.32.dist-info/RECORD

@@ -0,0 +1,7 @@
+gym_examples/__init__.py,sha256=EYrWJ3V1oe3cV6qzc6PCLGBpaidU210LmmACYpG1bt4,193
+gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
+gym_examples/envs/wsn_env.py,sha256=gZkGOQuMxtNO40vk-D1k90ed_iIdJdUQ3LFbpWkjxmU,14710
+gym_examples-2.0.32.dist-info/METADATA,sha256=IokXfCpcv9CpKNwR5aId5HN4I1UVNd5Of6E2nRy1adc,411
+gym_examples-2.0.32.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+gym_examples-2.0.32.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
+gym_examples-2.0.32.dist-info/RECORD,,

gym_examples-2.0.30.dist-info/RECORD

@@ -1,7 +0,0 @@
-gym_examples/__init__.py,sha256=uur4fOKfdZbOprSE_MqhnHFOEEgMbR2OFIQuRIJajoY,193
-gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
-gym_examples/envs/wsn_env.py,sha256=aKPaIgkMkNNynFuIvYD2ygW9R-0n_fRZKJPnj0KNT3M,13162
-gym_examples-2.0.30.dist-info/METADATA,sha256=X760BTR7bGZ1W19jrXFMsO4l1FueItkw7pEy7FNHqww,411
-gym_examples-2.0.30.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-gym_examples-2.0.30.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
-gym_examples-2.0.30.dist-info/RECORD,,