Rl-Unit1-CoLab / results.json
alanwsx's picture
First try of Unit 1 CoLab with PPO trained with 1 million steps
33ba93c verified
{"mean_reward": 246.43619689999997, "std_reward": 41.40578577102013, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2024-12-22T23:46:11.144783"}