| { | |
| "env_config": { | |
| "env_id": "LunarLander-v3", | |
| "env_kwargs": {}, | |
| "max_steps": null, | |
| "normalize_obs": false, | |
| "use_image": true, | |
| "vector_env_num": 6, | |
| "use_multi_processing": true, | |
| "image_shape": [ | |
| 84, | |
| 84 | |
| ], | |
| "frame_stack": 4, | |
| "frame_skip": 2, | |
| "training_render_mode": "rgb_array" | |
| }, | |
| "device": "mps", | |
| "learning_rate": 0.0001, | |
| "gamma": 0.99, | |
| "checkpoint_pathname": "", | |
| "max_grad_norm": 0.5, | |
| "log_interval": 100, | |
| "eval_episodes": 100, | |
| "eval_random_seed": 42, | |
| "eval_video_num": 10, | |
| "timesteps": 225000, | |
| "epsilon_schedule": { | |
| "_type": "ConstantSchedule", | |
| "_module": "practice.utils_for_coding.scheduler_utils", | |
| "value": 0.0 | |
| }, | |
| "replay_buffer_capacity": 0, | |
| "batch_size": 64, | |
| "train_interval": 1, | |
| "target_update_interval": 250, | |
| "update_start_step": 2000, | |
| "dqn_algorithm": "rainbow", | |
| "noisy_std": 0.5, | |
| "per_buffer_config": { | |
| "capacity": 135000, | |
| "n_step": 3, | |
| "gamma": 0.99, | |
| "use_uniform_sampling": true, | |
| "alpha": 0.6, | |
| "beta": 0.4, | |
| "beta_increment": 2.424242424242424e-06 | |
| }, | |
| "v_min": -300.0, | |
| "v_max": 300.0, | |
| "num_atoms": 51 | |
| } |