From b9dfcd0e8cf9b06a8815a6fe664a344948a7405a Mon Sep 17 00:00:00 2001
From: Joe251 <940621557@qq.com>
Date: Thu, 27 Dec 2018 22:33:06 +0800
Subject: [PATCH] Problem of max episode steps is fixed

---
 rl2/cartpole/random_search.py | 3 ++-
 rl2/cartpole/save_a_video.py  | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/rl2/cartpole/random_search.py b/rl2/cartpole/random_search.py
index 77ea36d6..1b16b2e3 100644
--- a/rl2/cartpole/random_search.py
+++ b/rl2/cartpole/random_search.py
@@ -19,7 +19,7 @@ def play_one_episode(env, params):
   done = False
   t = 0
 
-  while not done and t < 10000:
+  while not done:
     # env.render()
     t += 1
     action = get_action(observation, params)
@@ -58,6 +58,7 @@ def random_search(env):
 
 if __name__ == '__main__':
   env = gym.make('CartPole-v0')
+  env._max_episode_steps = 10000
   episode_lengths, params = random_search(env)
   plt.plot(episode_lengths)
   plt.show()
diff --git a/rl2/cartpole/save_a_video.py b/rl2/cartpole/save_a_video.py
index 31690c29..c423e3b7 100644
--- a/rl2/cartpole/save_a_video.py
+++ b/rl2/cartpole/save_a_video.py
@@ -20,7 +20,7 @@ def play_one_episode(env, params):
   done = False
   t = 0
 
-  while not done and t < 10000:
+  while not done:
     t += 1
     action = get_action(observation, params)
     observation, reward, done, info = env.step(action)
@@ -58,6 +58,7 @@ def random_search(env):
 
 if __name__ == '__main__':
   env = gym.make('CartPole-v0')
+  env._max_episode_steps = 10000
   episode_lengths, params = random_search(env)
   plt.plot(episode_lengths)
   plt.show()