kylefang/AI_for_video_games_demo
1
0
Fork 0
mirror of https://github.com/zhigang1992/AI_for_video_games_demo.git synced 2026-01-12 17:12:28 +08:00
Code Issues Packages Projects Releases Wiki Activity

Lock down gym version

Browse Source
This commit is contained in:
Zhigang Fang
2017-11-17 21:27:19 +08:00
parent eb1a697ff5
commit 0619789db1
3 changed files with 4 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
policy_iteration_demo.py
View File

@@ -71,7 +71,7 @@ def policy_iteration(env, gamma = 1.0):
if __name__ == '__main__': if __name__ == '__main__':
env_name = 'FrozenLake8x8-v0' env_name = 'FrozenLake8x8-v0'
env = gym.make(env_name) env = gym.make(env_name).unwrapped
optimal_policy = policy_iteration(env, gamma = 1.0) optimal_policy = policy_iteration(env, gamma = 1.0)
scores = evaluate_policy(env, optimal_policy, gamma = 1.0) scores = evaluate_policy(env, optimal_policy, gamma = 1.0)
print('Average scores = ', np.mean(scores)) print('Average scores = ', np.mean(scores))

2
requirements.txt Normal file
View File

@@ -0,0 +1,2 @@
gym[all]==0.9.4
numpy

2
value_iteration.py
View File

@@ -75,7 +75,7 @@ def value_iteration(env, gamma = 1.0):
if __name__ == '__main__': if __name__ == '__main__':
env_name = 'FrozenLake8x8-v0' env_name = 'FrozenLake8x8-v0'
gamma = 1.0 gamma = 1.0
env = gym.make(env_name) env = gym.make(env_name).unwrapped
optimal_v = value_iteration(env, gamma); optimal_v = value_iteration(env, gamma);
policy = extract_policy(optimal_v, gamma) policy = extract_policy(optimal_v, gamma)
policy_score = evaluate_policy(env, policy, gamma, n=1000) policy_score = evaluate_policy(env, policy, gamma, n=1000)