| { | |
| "env_config": { | |
| "env_id": "Walker2d-v5", | |
| "env_kwargs": {}, | |
| "max_steps": null, | |
| "use_image": false, | |
| "vector_env_num": 6, | |
| "use_multi_processing": true, | |
| "image_shape": null, | |
| "frame_stack": 1, | |
| "frame_skip": 1, | |
| "training_render_mode": null | |
| }, | |
| "device": "mps", | |
| "learning_rate": 0.0002, | |
| "gamma": 0.995, | |
| "checkpoint_pathname": "", | |
| "eval_episodes": 50, | |
| "eval_random_seed": 42, | |
| "eval_video_num": 10, | |
| "total_steps": 2400000, | |
| "hidden_sizes": [ | |
| 400, | |
| 300 | |
| ], | |
| "critic_lr": 0.0003, | |
| "replay_buffer_capacity": 1440000, | |
| "batch_size": 256, | |
| "update_start_step": 10000, | |
| "policy_delay": 2, | |
| "policy_noise": 0.2, | |
| "noise_clip": 0.5, | |
| "exploration_noise": { | |
| "_type": "LinearSchedule", | |
| "_module": "practice.utils_for_coding.scheduler_utils", | |
| "start_e": 0.4, | |
| "end_e": 0.05, | |
| "duration": 200000 | |
| }, | |
| "max_action": 1.0, | |
| "tau": 0.05, | |
| "max_grad_norm": 0.5, | |
| "smooth_l1_loss_beta": null | |
| } |