{"hidden_size": 16, "n_training_episodes": 10000, "n_evaluation_episodes": 10, "max_t": 450, "gamma": 1.0, "lr": 0.01, "env_id": "CartPole-v1", "states_size": 4, "actions_size": 2, "env": "<TimeLimit<OrderEnforcing<PassiveEnvChecker<CartPoleEnv<CartPole-v1>>>>>"}