22 lines
436 B
Python
22 lines
436 B
Python
import gym
|
|
|
|
from baselines import deepq
|
|
|
|
|
|
def main():
|
|
env = gym.make("CartPole-v0")
|
|
act = deepq.load("cartpole_model.pkl")
|
|
|
|
while True:
|
|
obs, done = env.reset(), False
|
|
episode_rew = 0
|
|
while not done:
|
|
env.render()
|
|
obs, rew, done, _ = env.step(act(obs[None])[0])
|
|
episode_rew += rew
|
|
print("Episode reward", episode_rew)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
main()
|