Reinforcement Learning: Gym's Built-in & third-party environments

Author: DevOpsThinh

README.md

@@ -33,6 +33,15 @@ ___
 ___
 
 ### Illustrated results
+
+#### Gym's Built-in & third-party environments
+- Lunar Lander<br>
+<img src="ml/rl_in_robotics/images/lunar_lander.gif" height="480"/> <br>
+- Mountain Car<br>
+<img src="ml/rl_in_robotics/images/mountain_car.gif" height="480"/> <br>
+- Phoenix<br>
+<img src="ml/rl_in_robotics/images/phoenix.gif" height="480"/> <br>
+
 #### How reinforcement learning helps a robot balance a pole
 - Random movements cannot keep the balance for a long time<br>
 <img src="ml/rl_in_robotics/images/cart-pole-rl.gif" height="480"/> <br>

ml/rl_in_robotics/README.md

@@ -0,0 +1,7 @@
+## Packages Installation for Python 3.8+
+
+- pip install gym==0.19.0
+- pip install gym[atari]==0.19.0 or <br> pip install atari-py==0.2.6, then [Download ale_c.dll](https://stackoverflow.com/questions/63080326/could-not-find-module-atari-py-ale-interface-ale-c-dll-or-one-of-its-dependenc) <br>
+- pip install pyglet==1.5.27
+- pip install swig
+- pip install ufal.pybox2d==2.3.10

ml/rl_in_robotics/cart-pole_balancing_rl.py renamed to ml/rl_in_robotics/cart_pole/cart_pole_balancing_rl.py

@@ -35,10 +35,7 @@
         sleep(.01)
         if done:
             print(f'Episode {i} reward: {reward_sum}')
-            sleep(.01)
-            if done:
-                print(f'Episode {i} reward: {reward_sum}')
-                sleep(1)
-                break
+            sleep(1)
+            break
 
 env.close()

ml/rl_in_robotics/cart_pole_rl.py renamed to ml/rl_in_robotics/cart_pole/cart_pole_rl.py

@@ -38,10 +38,7 @@
         sleep(.01)
         if done:
             print(f'Episode {i} reward: {reward_sum}')
-            sleep(.01)
-            if done:
-                print(f'Episode {i} reward: {reward_sum}')
-                sleep(1)
-                break
+            sleep(1)
+            break
 
 env.close()

ml/rl_in_robotics/gym_environments/lunar_lander.py

@@ -0,0 +1,18 @@
+# Learner: Nguyen Truong Thinh
+# Contact me: [email protected] || +84393280504
+#
+# Topic: Reinforcement Learning (RL): RL in robotics
+#           Gym environments: Lunar Lander Reinforcement Learning
+#                   The agent's goal is to land the lander between yellow flags.
+
+from ml.rl_in_robotics.utility import gym_rl_tasks, init_environment
+
+# Hyperparameters we can adjust
+EPISODES = 10
+
+env = init_environment("LunarLander-v2")
+
+random_action = env.action_space.sample()
+
+gym_rl_tasks(env, EPISODES, random_action)
+

ml/rl_in_robotics/gym_environments/phoenix.py

@@ -0,0 +1,18 @@
+# Learner: Nguyen Truong Thinh
+# Contact me: [email protected] || +84393280504
+#
+# Topic: Reinforcement Learning (RL): RL in robotics
+#           Gym environments: Phoenix Reinforcement Learning
+#                   This environment is one of the legendary Atari video game series.
+#                   In this game, you have to maximize your score.
+
+from ml.rl_in_robotics.utility import gym_rl_tasks, init_environment
+
+# Hyperparameters we can adjust
+EPISODES = 10
+
+env = init_environment("Phoenix-v0")
+
+random_action = env.action_space.sample()
+
+gym_rl_tasks(env, EPISODES, random_action)

ml/rl_in_robotics/gym_environments/right_mountain_car.py

@@ -0,0 +1,18 @@
+# Learner: Nguyen Truong Thinh
+# Contact me: [email protected] || +84393280504
+#
+# Topic: Reinforcement Learning (RL): RL in robotics
+#           Gym environments: Mountain Car Reinforcement Learning
+#                   A car is positioned between two mountains.
+#                   The goal is to reach the mountain on the right.
+
+from ml.rl_in_robotics.utility import gym_rl_tasks, init_environment
+
+# Hyperparameters we can adjust
+EPISODES = 30
+
+env = init_environment("MountainCar-v0")
+
+random_action = env.action_space.sample()
+
+gym_rl_tasks(env, EPISODES, random_action)