gym-examples 3.0.767__py3-none-any.whl → 3.0.768__py3-none-any.whl

gym_examples/__init__.py

@@ -5,4 +5,4 @@ register(
      entry_point="gym_examples.envs:WSNRoutingEnv",
 )
 
-__version__ = "3.0.767"
+__version__ = "3.0.768"

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 = [0] * input_dim
         rewards = [reward_init] * self.n_sensors
         dones = [False] * self.n_sensors
         for i, action in enumerate(actions):
@@ -159,7 +159,7 @@ class WSNRoutingEnv(gym.Env):
             
             if (action == i):
                 continue  # Skip if sensor tries to transmit data to itself
-
+            
             if action == self.n_sensors:
                 if self.distance_to_base[i] > self.coverage_radius:
                     continue  # Skip if the distance to the base station is greater than the coverage radius
@@ -492,7 +492,7 @@ class WSNRoutingEnv(gym.Env):
         query_vector = np.array([max_reward] * d) # Basically, the target is the base station
 
         # Similarities between the query vector and the rewards
-        similarities = [np.dot(query_vector, reward) for reward in rewards]
+        similarities = [np.dot(query_vector, reward) for reward in rewards if isinstance(reward, np.ndarray)]
 
         # Similarities scaling
         similarities = [similarity / np.sqrt(d) for similarity in similarities]
@@ -502,7 +502,7 @@ class WSNRoutingEnv(gym.Env):
         attention_weights = [np.exp(similarity) / denominator for similarity in similarities]
 
         # Weighted sum of the rewards
-        final_reward = sum([attention_weight * reward for attention_weight, reward in zip(attention_weights, rewards)])
+        final_reward = sum([attention_weight * reward for attention_weight, reward in zip(attention_weights, rewards) if isinstance(reward, np.ndarray)])
 
         return np.sum(final_reward)
 

{gym_examples-3.0.767.dist-info → gym_examples-3.0.768.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gym-examples
-Version: 3.0.767
+Version: 3.0.768
 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-3.0.768.dist-info/RECORD

@@ -0,0 +1,7 @@
+gym_examples/__init__.py,sha256=ErBo6hRVSUgQvxu0zaJqLRmZM9LTYXQ_1HIgV4bg01o,166
+gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
+gym_examples/envs/wsn_env.py,sha256=baJSRHpxNnasenWc-lGGjeHvXkJQaSLSpr_pSP9SrmQ,26864
+gym_examples-3.0.768.dist-info/METADATA,sha256=U9TYdpAF3BmSEktZwNYaDSjCV3bjKezIUbAMsQBgfGs,412
+gym_examples-3.0.768.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+gym_examples-3.0.768.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
+gym_examples-3.0.768.dist-info/RECORD,,

gym_examples-3.0.767.dist-info/RECORD

@@ -1,7 +0,0 @@
-gym_examples/__init__.py,sha256=lI2A534ndKHi9RQYvkCYTX1c8O0mQr5rLLLQuRzu5Qw,166
-gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
-gym_examples/envs/wsn_env.py,sha256=QX9nNxbq87fRr0Dtfwe3aPtTmtiCJtNHixJLfycR1mo,26794
-gym_examples-3.0.767.dist-info/METADATA,sha256=KSj6oEzWvhbdwcJxk6M1V_OxbwExmam4LyXCHUbrWZ8,412
-gym_examples-3.0.767.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-gym_examples-3.0.767.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
-gym_examples-3.0.767.dist-info/RECORD,,