gym-examples 3.0.769__py3-none-any.whl → 3.0.771__py3-none-any.whl

gym_examples/__init__.py

@@ -5,4 +5,4 @@ register(
      entry_point="gym_examples.envs:WSNRoutingEnv",
 )
 
-__version__ = "3.0.769"
+__version__ = "3.0.771"

gym_examples/envs/wsn_env.py

@@ -150,7 +150,7 @@ class WSNRoutingEnv(gym.Env):
         self.number_of_steps += 1
         self.steps += 1 
         # rewards = [-max_reward] * self.n_sensors
-        reward_init = np.array([0] * input_dim)
+        reward_init = np.array([- max_reward] * input_dim)
         rewards = [reward_init] * self.n_sensors
         dones = [False] * self.n_sensors
         for i, action in enumerate(actions):
@@ -219,6 +219,8 @@ class WSNRoutingEnv(gym.Env):
                 
                 self.number_of_packets[i] = 0 # Reset the number of packets of the sensor i
         
+            # rewards[i] = self.compute_attention_rewards(rewards[i]) # Compute the attention-based reward
+
         # Integrate the mobility of the sensors
         # self.integrate_mobility() 
 
@@ -231,6 +233,7 @@ class WSNRoutingEnv(gym.Env):
 
         # rewards = [reward.item() if isinstance(reward, torch.Tensor) else reward for reward in rewards] # Convert the reward to a float
         rewards = self.compute_attention_reward(rewards) # Compute the attention-based reward
+        # rewards = np.sum(rewards) # Sum the rewards of all the sensors
 
         for i in range(self.n_sensors):
             if not dones[i]:
@@ -472,8 +475,7 @@ class WSNRoutingEnv(gym.Env):
     #     Compute the attention-based rewards
     #     '''
     #     rewards_i = torch.tensor(reward, dtype=torch.double)
-    #     rewards_i = rewards_i.unsqueeze(0) # Add batch dimension
-    #     return net(rewards_i).item()
+    #     return net(rewards_i)
 
 
     def compute_attention_reward(self, rewards):

{gym_examples-3.0.769.dist-info → gym_examples-3.0.771.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gym-examples
-Version: 3.0.769
+Version: 3.0.771
 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-3.0.771.dist-info/RECORD

@@ -0,0 +1,7 @@
+gym_examples/__init__.py,sha256=RbdE0YwrLSAqeAvXUmj2eOfSVERmLe_xP34l5_GGTFc,166
+gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
+gym_examples/envs/wsn_env.py,sha256=CEI6uM78EVY5QEXjslxfaTXDP-wTSklrM9yghckQ-vk,26926
+gym_examples-3.0.771.dist-info/METADATA,sha256=dzdtTI0gYHEF8BR04nWmRW-xtF235JsM9Q7GyIbxj7U,412
+gym_examples-3.0.771.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+gym_examples-3.0.771.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
+gym_examples-3.0.771.dist-info/RECORD,,

gym_examples-3.0.769.dist-info/RECORD

@@ -1,7 +0,0 @@
-gym_examples/__init__.py,sha256=JrDpTNbLnzE4_62ux2BYcurQHhbfV4ZjpaK9TfNmgvM,166
-gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
-gym_examples/envs/wsn_env.py,sha256=O2DXjVkYubXn5aLNeCf2AwQ1FUZlwp8R2xYCR1MDqCk,26806
-gym_examples-3.0.769.dist-info/METADATA,sha256=6oumqB5XQ0yhGHDK7G4E5yj2JeK0TxSqFqb3XUsUjHw,412
-gym_examples-3.0.769.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-gym_examples-3.0.769.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
-gym_examples-3.0.769.dist-info/RECORD,,