gym-examples 3.0.70__py3-none-any.whl → 3.0.71__py3-none-any.whl

gym_examples/__init__.py

@@ -6,4 +6,4 @@ register(
      max_episode_steps=50,
 )
 
-__version__ = "3.0.70"
+__version__ = "3.0.71"

gym_examples/envs/wsn_env.py

@@ -88,7 +88,7 @@ class WSNRoutingEnv(gym.Env):
 
     def step(self, actions):
         self.steps += 1
-        rewards = [0] * self.n_sensors
+        rewards = [-1] * self.n_sensors
         dones = [False] * self.n_sensors
         for i, action in enumerate(actions):
             if action not in range(self.n_sensors + 1):
@@ -165,8 +165,6 @@ class WSNRoutingEnv(gym.Env):
 
         self.get_metrics()
 
-        # rewards = [r / len(rewards) for r in rewards]
-
         return self._get_obs(), rewards, dones, {}
 
     def _get_obs(self):
@@ -224,7 +222,8 @@ class WSNRoutingEnv(gym.Env):
         # Normalize the angle
         normalized_angle = abs(angle) / np.pi
 
-        return np.clip(1 - normalized_angle, 0, 1)
+        # return np.clip(1 - normalized_angle, 0, 1)
+        return np.clip(- normalized_angle, -1, 1)
     
     def compute_reward_distance(self, i, action):
         '''
@@ -237,7 +236,8 @@ class WSNRoutingEnv(gym.Env):
         # Normalize the distance to the next hop
         normalized_distance_to_next_hop = distance / self.coverage_radius
 
-        return np.clip(1 - normalized_distance_to_next_hop, 0, 1)
+        # return np.clip(1 - normalized_distance_to_next_hop, 0, 1)
+        return np.clip(-normalized_distance_to_next_hop, -1, 1)
 
     def compute_reward_consumption_energy(self, i, action):
         '''
@@ -258,8 +258,9 @@ class WSNRoutingEnv(gym.Env):
         max_total_energy = max_transmission_energy + max_reception_energy
         normalized_total_energy = total_energy / (max_total_energy + self.epsilon)
 
-        return np.clip(1 - normalized_total_energy, 0, 1)
-    
+        # return np.clip(1 - normalized_total_energy, 0, 1)
+        return np.clip(- normalized_total_energy, -1, 1)
+
     def compute_reward_dispersion_remaining_energy(self):
         '''
         Compute the reward based on the standard deviation of the remaining energy
@@ -269,8 +270,9 @@ class WSNRoutingEnv(gym.Env):
         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 + self.epsilon)
 
-        return np.clip(1 - normalized_dispersion_remaining_energy, 0, 1)
-    
+        # return np.clip(1 - normalized_dispersion_remaining_energy, 0, 1)
+        return np.clip(- normalized_dispersion_remaining_energy, -1, 1)
+
     def compute_reward_number_of_packets(self, action):
         '''
         Compute the reward based on the number of packets of the receiver
@@ -281,7 +283,8 @@ class WSNRoutingEnv(gym.Env):
         else: 
             normalized_number_of_packets = self.number_of_packets[action] / (max_number_of_packets + self.epsilon)
 
-        return np.clip(1 - normalized_number_of_packets, 0, 1)
+        # return np.clip(1 - normalized_number_of_packets, 0, 1)
+        return np.clip(- normalized_number_of_packets, -1, 1)
 
     def compute_individual_rewards(self, i, action):
         '''
@@ -315,7 +318,8 @@ class WSNRoutingEnv(gym.Env):
         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 + self.epsilon)
 
-        return np.clip(1 - normalized_dispersion_remaining_energy, 0, 1)
+        # return np.clip(1 - normalized_dispersion_remaining_energy, 0, 1)
+        return np.clip(- normalized_dispersion_remaining_energy, -1, 1)
     
     def network_reward_consumption_energy(self):
         '''
@@ -326,7 +330,8 @@ class WSNRoutingEnv(gym.Env):
         max_total_energy = self.n_sensors * initial_energy
         normalized_total_energy = total_energy / (max_total_energy + self.epsilon)
 
-        return np.clip(1 - normalized_total_energy, 0, 1)
+        # return np.clip(1 - normalized_total_energy, 0, 1)
+        return np.clip(- normalized_total_energy, -1, 1)
     
     def compute_reward_packet_delivery_ratio(self):
         '''
@@ -343,7 +348,8 @@ class WSNRoutingEnv(gym.Env):
         max_latency = self.n_sensors * self.steps
         normalized_latency = self.total_latency / (max_latency + self.epsilon)
 
-        return np.clip(1 - normalized_latency, 0, 1)
+        # return np.clip(1 - normalized_latency, 0, 1)
+        return np.clip(- normalized_latency, -1, 1)
 
     def compute_reward_network_throughput(self):
         '''

{gym_examples-3.0.70.dist-info → gym_examples-3.0.71.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gym-examples
-Version: 3.0.70
+Version: 3.0.71
 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-3.0.71.dist-info/RECORD

@@ -0,0 +1,7 @@
+gym_examples/__init__.py,sha256=g7xElPNRTjXJyIh9gGRHxZuNWJ8yoQ4m9wG2vxXXR8U,193
+gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
+gym_examples/envs/wsn_env.py,sha256=6ozl-jBiWyPwRaUvgMAshbydZn56lg77VGqYI8SqT2c,20451
+gym_examples-3.0.71.dist-info/METADATA,sha256=O34q71LQmiTyPg53ntHDZAKanPvL9Xjd2IPpJKiASok,411
+gym_examples-3.0.71.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+gym_examples-3.0.71.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
+gym_examples-3.0.71.dist-info/RECORD,,

gym_examples-3.0.70.dist-info/RECORD

@@ -1,7 +0,0 @@
-gym_examples/__init__.py,sha256=90AkOI4OSX_jWRjyWz1Qwg43aeQY7I_e5zr_0JEf9-o,193
-gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
-gym_examples/envs/wsn_env.py,sha256=BNBpC74P-CK4F1rA59PTgYTUjMyGZclmJmDE4D4YBuA,20007
-gym_examples-3.0.70.dist-info/METADATA,sha256=fPNX7FRPRX9T1KVRK2E14UEUBVgwbhe1EQGd5sqU4O0,411
-gym_examples-3.0.70.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-gym_examples-3.0.70.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
-gym_examples-3.0.70.dist-info/RECORD,,