{ "env_config": { "env_id": "MountainCar-v0", "env_kwargs": {}, "max_steps": null, "normalize_obs": true, "use_image": false, "vector_env_num": null, "use_multi_processing": false, "image_shape": null, "frame_stack": 1, "frame_skip": 1, "training_render_mode": null }, "device": "cpu", "learning_rate": 0.003, "gamma": 0.999, "checkpoint_pathname": "", "max_grad_norm": 0.5, "log_interval": 100, "eval_episodes": 20, "eval_random_seed": 42, "eval_video_num": 10, "total_steps": 200000, "lr_gamma": 0.99, "hidden_sizes": [ 32, 32 ], "baseline": { "_type": "ConstantBaseline", "_module": "practice.utils_for_coding.baseline_utils", "_decay": 0.9, "_baseline_value": -31.932100605853133, "_baseline_initialized": true }, "entropy_coef": { "_type": "LinearSchedule", "_module": "practice.utils_for_coding.scheduler_utils", "_start_e": 0.01, "_end_e": 0.001, "_duration": 100000, "_start_t": 0 }, "reward_configs": [ { "rnd_config": { "beta": { "_type": "LinearSchedule", "_module": "practice.utils_for_coding.scheduler_utils", "_start_e": 0.005, "_end_e": 0.001, "_duration": 160000, "_start_t": 0 }, "normalize": true, "device": "cpu", "max_reward": 2 }, "obs_dim": 2, "output_dim": 32, "hidden_sizes": [ 32, 32 ], "learning_rate": 0.0005 }, { "beta": { "_type": "LinearSchedule", "_module": "practice.utils_for_coding.scheduler_utils", "_start_e": 5.0, "_end_e": 1.0, "_duration": 160000, "_start_t": 0 }, "goal_position": null } ] }