gym-examples 3.0.128__py3-none-any.whl → 3.0.130__py3-none-any.whl

gym_examples/__init__.py

@@ -6,4 +6,4 @@ register(
      max_episode_steps=50,
 )
 
-__version__ = "3.0.128"
+__version__ = "3.0.130"

gym_examples/envs/wsn_env.py

@@ -59,33 +59,10 @@ class WSNRoutingEnv(gym.Env):
         # Initialize the position of the sensors randomly
 
         # Define observation space
-        # self.observation_space = Tuple(
-        #     tuple([self._get_observation_space() for _ in range(self.n_sensors)])
-        # )
-        # self.action_space = Tuple(tuple([Discrete(self.n_sensors + 1)] * self.n_agents))
-
-        # Define the flattened observation space
-        single_sensor_obs_space = self._get_observation_space()
-        obs_low = np.concatenate([
-            single_sensor_obs_space['remaining_energy'].low,
-            single_sensor_obs_space['consumption_energy'].low,
-            single_sensor_obs_space['sensor_positions'].low,
-            single_sensor_obs_space['number_of_packets'].low
-        ])
-        obs_high = np.concatenate([
-            single_sensor_obs_space['remaining_energy'].high,
-            single_sensor_obs_space['consumption_energy'].high,
-            single_sensor_obs_space['sensor_positions'].high,
-            single_sensor_obs_space['number_of_packets'].high
-        ])
-        self.observation_space = Box(
-            low=np.tile(obs_low, self.n_sensors),
-            high=np.tile(obs_high, self.n_sensors),
-            dtype=np.float32
+        self.observation_space = Tuple(
+            tuple([self._get_observation_space() for _ in range(self.n_sensors)])
         )
-
-        # Define the flattened action space
-        self.action_space = MultiDiscrete([self.n_sensors + 1] * self.n_agents)
+        self.action_space = Tuple(tuple([Discrete(self.n_sensors + 1)] * self.n_agents))
 
         self.reset()        
 
@@ -196,7 +173,6 @@ class WSNRoutingEnv(gym.Env):
         return self._get_obs(), rewards, dones, {}
 
     # def _get_obs(self):
-
     #     return [{'remaining_energy': np.array([e]), 
     #              'consumption_energy': np.array([initial_energy - e]),
     #              'sensor_positions': p,
@@ -204,19 +180,18 @@ class WSNRoutingEnv(gym.Env):
     #             } for e, p, d in zip(self.remaining_energy, self.sensor_positions, self.number_of_packets)]
 
     def _get_obs(self):
-        # Structured observation
-        structured_obs = [{'remaining_energy': np.array([e]), 
-                           'consumption_energy': np.array([initial_energy - e]),
-                           'sensor_positions': p,
-                           'number_of_packets': np.array([d])
-                          } for e, p, d in zip(self.remaining_energy, self.sensor_positions, self.number_of_packets)]
-
-        # Flattened observation for compatibility with Stable Baselines3
-        flat_obs = np.concatenate([
-            np.concatenate([obs['remaining_energy'], obs['consumption_energy'], obs['sensor_positions'], obs['number_of_packets']])
-            for obs in structured_obs
-        ])
-        return flat_obs
+        remaining_energy = np.array([e for e, _, _, _ in zip(self.remaining_energy, self.sensor_positions, self.number_of_packets)])
+        consumption_energy = np.array([initial_energy - e for e, _, _, _ in zip(self.remaining_energy, self.sensor_positions, self.number_of_packets)])
+        sensor_positions = np.stack(self.sensor_positions, axis=0)
+        number_of_packets = np.array([d for _, _, _, d in zip(self.remaining_energy, self.sensor_positions, self.number_of_packets)])
+
+        return {
+            'remaining_energy': remaining_energy,
+            'consumption_energy': consumption_energy,
+            'sensor_positions': sensor_positions,
+            'number_of_packets': number_of_packets
+        }
+
 
     def _get_observation_space(self):
         return Dict({

{gym_examples-3.0.128.dist-info → gym_examples-3.0.130.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gym-examples
-Version: 3.0.128
+Version: 3.0.130
 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-3.0.130.dist-info/RECORD

@@ -0,0 +1,7 @@
+gym_examples/__init__.py,sha256=ephqHG1V21jR1hU69gk7ByNV2lo-xcz__sSSxL0TRiQ,194
+gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
+gym_examples/envs/wsn_env.py,sha256=jPME2gvkDL7BfMqYgjpAXna-mTAv3_81HlPs-U5FnTM,22286
+gym_examples-3.0.130.dist-info/METADATA,sha256=28_KBJf-MOIyD0HgbmzMvu1UlAcH5s-o0az8Qd1pN1Y,412
+gym_examples-3.0.130.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+gym_examples-3.0.130.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
+gym_examples-3.0.130.dist-info/RECORD,,

gym_examples-3.0.128.dist-info/RECORD

@@ -1,7 +0,0 @@
-gym_examples/__init__.py,sha256=HtlSQQsIPVhRkM90RZRf4j2KCKHhI4jQzcBnLzxNT1U,194
-gym_examples/envs/__init__.py,sha256=lgMe4pyOuUTgTBUddM0iwMlETsYTwFShny6ifm8PGM8,53
-gym_examples/envs/wsn_env.py,sha256=cPfhzLLoAMoBonuR7PdgS-OjU9e1AbaJlgRo4AxHTqY,23329
-gym_examples-3.0.128.dist-info/METADATA,sha256=CCcUSZp8Tyavj_Gxy9eWmZH1XJgvrMbItHuVR6BlRXM,412
-gym_examples-3.0.128.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-gym_examples-3.0.128.dist-info/top_level.txt,sha256=rJRksoAF32M6lTLBEwYzRdo4PgtejceaNnnZ3HeY_Rk,13
-gym_examples-3.0.128.dist-info/RECORD,,