Skip to main content

Prabhasak's group workspace

Timestamps visible
2020-09-24 12:39:57
Episode length: 500.00 +/- 0.00
2020-09-24 12:49:21
Eval num_timesteps=899584, episode_reward=500.00 +/- 0.00
2020-09-24 12:49:21
Episode length: 500.00 +/- 0.00
2020-09-24 12:58:42
Eval num_timesteps=949760, episode_reward=500.00 +/- 0.00
2020-09-24 12:58:42
Episode length: 500.00 +/- 0.00
2020-09-24 13:07:44
Eval num_timesteps=999936, episode_reward=500.00 +/- 0.00
2020-09-24 13:07:44
Episode length: 500.00 +/- 0.00
2020-09-24 13:07:49
Choosing best saved model throughout training, instead of model available at end of training (EvalCallback enabled!)
2020-09-24 13:07:51
Trained CartPole-v1 with gail. Some stats:
2020-09-24 13:07:51

2020-09-24 13:09:27
[99]/100 successful episodes
2020-09-24 13:09:27

2020-09-24 13:09:27
Mean return:  495.17
2020-09-24 13:09:27
Std return:  48.05789
2020-09-24 13:09:27
Max return:  [500.]
2020-09-24 13:09:27
Min return:  [17.]
2020-09-24 13:09:27
Mean episode len:  495.0
2020-09-24 13:09:27
Optimal policy found!
2020-09-24 13:09:27