gym-examples 3.0.178__py3-none-any.whl → 3.0.180__py3-none-any.whl

gym_examples/__init__.py

@@ -6,4 +6,4 @@ register(
      max_episode_steps=50,
 )
 
-__version__ = "3.0.178"
+__version__ = "3.0.180"

gym_examples/envs/wsn_env.py

@@ -55,12 +55,8 @@ class WSNRoutingEnv(gym.Env):
     def __init__(self, n_sensors = 20, coverage_radius=(upper_bound - lower_bound)/4):
 
         super(WSNRoutingEnv, self).__init__()
-        self.log_done = None # Log the done status of the environment
-        self.log_action = None # Log the action taken by the agent
-        self.log_steps = None # Log the number of steps taken by the agent
 
         # Initialize list of episode metrics
-        episode_return = None
         self.number_of_steps = 0 # Total number of steps taken by the agent since the beginning of the training 
         self.episode_returns = []
         self.episode_std_remaining_energy = []
@@ -91,10 +87,8 @@ class WSNRoutingEnv(gym.Env):
 
     def reset(self):
 
-        episode_return = 0
-
         if self.episode_count > 1 and os.getenv('PRINT_STATS') == 'True':
-            self.episode_returns.append(episode_return)
+            self.episode_returns.append(self.episode_return)
             self.episode_std_remaining_energy.append(np.std(self.remaining_energy))
             self.episode_mean_remaining_energy.append(np.mean(self.remaining_energy))
             self.episode_total_consumption_energy.append(np.sum(initial_energy - self.remaining_energy))
@@ -102,11 +96,10 @@ class WSNRoutingEnv(gym.Env):
             self.episode_packet_delivery_ratio.append(self.packet_delivery_ratio)
             self.episode_network_lifetime.append(self.network_lifetime)
             self.episode_average_latency.append(self.average_latency)
-            print(f"Episode: {self.episode_count}")
-            print(f"This episode ends with # Steps: {self.log_steps}")
-            print(f"This episode ends with Done: {self.log_done}")
-            print(f"This episode ends with Action: {self.log_action}")            
-            print(self.get_metrics())                
+            print(f"Episode: {self.episode_count}")           
+            print(self.get_metrics())  
+
+        self.episode_return = 0              
         self.sensor_positions = np.random.rand(self.n_sensors, 2) * (upper_bound - lower_bound) + lower_bound
         self.distance_to_base = np.linalg.norm(self.sensor_positions - base_station_position, axis=1)
         self.remaining_energy = np.ones(self.n_sensors) * initial_energy
@@ -131,9 +124,7 @@ class WSNRoutingEnv(gym.Env):
 
     def step(self, actions):
         actions = [actions[i] for i in range(self.n_agents)] # We want to go back from the MultiDiscrete action space to a tuple of tuple of Discrete action spaces
-        self.steps += 1
-        if os.getenv('PRINT_STATS') == 'True':
-            self.number_of_steps += 1 
+        self.steps += 1 
         rewards = [0] * self.n_sensors
         dones = [False] * self.n_sensors
         for i, action in enumerate(actions):
@@ -217,14 +208,12 @@ class WSNRoutingEnv(gym.Env):
         rewards = np.mean(rewards)  # Average the rewards of all agents
         # print(f"Step: {self.steps}, Rewards: {rewards}, Done: {dones}")
         dones = all(dones)  # Done if all agents are done
-        self.log_done = dones
-        self.log_action = actions
-        self.log_steps = self.steps
-        episode_return += rewards
-
+        
         if os.getenv('PRINT_STATS') == 'True': # We are trying to extract only the statistics for the PPO algorithm
+            self.number_of_steps += 1
+            self.episode_return += rewards
             if self.number_of_steps >= num_timesteps:
-                self.episode_returns.append(episode_return)
+                self.episode_returns.append(self.episode_return)
                 self.episode_std_remaining_energy.append(np.std(self.remaining_energy))
                 self.episode_mean_remaining_energy.append(np.mean(self.remaining_energy))
                 self.episode_total_consumption_energy.append(np.sum(initial_energy - self.remaining_energy))

{gym_examples-3.0.178.dist-info → gym_examples-3.0.180.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gym-examples
-Version: 3.0.178
+Version: 3.0.180
 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-3.0.180.dist-info/RECORD

@@ -0,0 +1,7 @@
+gym_examples/__init__.py,sha256=u596NUGuugLKs4xG38CdXc1ChDTjdSFE4qwRq07dSco,194
+gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
+gym_examples/envs/wsn_env.py,sha256=xXwNW7X-pFQT0T_vJwI7WIkoAhUhtmz7nTEokL9d30Q,25660
+gym_examples-3.0.180.dist-info/METADATA,sha256=esLOv32LCP3zP5pOa-zczAQTsz0ZSha7VqVAL0pcsNg,412
+gym_examples-3.0.180.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+gym_examples-3.0.180.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
+gym_examples-3.0.180.dist-info/RECORD,,

gym_examples-3.0.178.dist-info/RECORD

@@ -1,7 +0,0 @@
-gym_examples/__init__.py,sha256=dKB7fxf-b5wOgFhNlcNdhFqGY0_sagdM3FVbJyd2xBM,194
-gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
-gym_examples/envs/wsn_env.py,sha256=MsDYLY0-6ngqgziKcJ5BET701GWX6OU1qoh3FMTezcs,26238
-gym_examples-3.0.178.dist-info/METADATA,sha256=zRHR4aBsr9tc5dwMeqysMfhLm8823FOnSCj7tph9DaM,412
-gym_examples-3.0.178.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-gym_examples-3.0.178.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
-gym_examples-3.0.178.dist-info/RECORD,,