Added a finite horizon=100 env registered in Gym.

Author: RaghuSpaceRajan

example.py

@@ -229,4 +229,7 @@ def mujoco_wrapper_example():
     import mdp_playground
     import gym
     gym.make('RLToy-v0')
-    gym.make('RLToy-v0', **{'state_space_size':8, 'action_space_size':8, 'state_space_type':'discrete', 'action_space_type':'discrete', 'terminal_state_density':0.25, 'maximally_connected': True})
+    env = gym.make('RLToyFiniteHorizon-v0', **{'state_space_size':8, 'action_space_size':8, 'state_space_type':'discrete', 'action_space_type':'discrete', 'terminal_state_density':0.25, 'maximally_connected': True})
+    env.reset()
+    for i in range(10):
+        print(env.step(env.action_space.sample()))

mdp_playground/__init__.py

@@ -4,3 +4,9 @@
     id='RLToy-v0',
     entry_point='mdp_playground.envs:RLToyEnv',
 )
+
+register(
+    id='RLToyFiniteHorizon-v0',
+    entry_point='mdp_playground.envs:RLToyEnv',
+    max_episode_steps=100,
+)