{"h_size": 64, "n_training_episodes": 2000, "n_evaluation_episodes": 10, "max_t": 500, "gamma": 1, "lr": 0.001, "env_id": "LunarLander-v2", "state_space": 8, "action_space": 4}