| exp_config = { | |
| 'env': { | |
| 'manager': { | |
| 'episode_num': float("inf"), | |
| 'max_retry': 1, | |
| 'retry_type': 'reset', | |
| 'auto_reset': True, | |
| 'step_timeout': None, | |
| 'reset_timeout': None, | |
| 'retry_waiting_time': 0.1, | |
| 'cfg_type': 'BaseEnvManagerDict' | |
| }, | |
| 'stop_value': 240, | |
| 'n_evaluator_episode': 8, | |
| 'env_id': 'LunarLanderContinuous-v2', | |
| 'collector_env_num': 4, | |
| 'evaluator_env_num': 8, | |
| 'act_scale': True | |
| }, | |
| 'policy': { | |
| 'model': { | |
| 'twin_critic': True, | |
| 'obs_shape': 8, | |
| 'action_shape': 2, | |
| 'action_space': 'regression' | |
| }, | |
| 'learn': { | |
| 'learner': { | |
| 'train_iterations': 1000000000, | |
| 'dataloader': { | |
| 'num_workers': 0 | |
| }, | |
| 'log_policy': True, | |
| 'hook': { | |
| 'load_ckpt_before_run': '', | |
| 'log_show_after_iter': 100, | |
| 'save_ckpt_after_iter': 10000, | |
| 'save_ckpt_after_run': True | |
| }, | |
| 'cfg_type': 'BaseLearnerDict' | |
| }, | |
| 'update_per_collect': 256, | |
| 'batch_size': 256, | |
| 'learning_rate_actor': 0.0003, | |
| 'learning_rate_critic': 0.001, | |
| 'ignore_done': False, | |
| 'target_theta': 0.005, | |
| 'discount_factor': 0.99, | |
| 'actor_update_freq': 2, | |
| 'noise': True, | |
| 'noise_sigma': 0.1, | |
| 'noise_range': { | |
| 'min': -0.5, | |
| 'max': 0.5 | |
| } | |
| }, | |
| 'collect': { | |
| 'collector': {}, | |
| 'unroll_len': 1, | |
| 'noise_sigma': 0.1, | |
| 'n_sample': 256 | |
| }, | |
| 'eval': { | |
| 'evaluator': { | |
| 'eval_freq': 1000, | |
| 'render': { | |
| 'render_freq': -1, | |
| 'mode': 'train_iter' | |
| }, | |
| 'figure_path': None, | |
| 'cfg_type': 'InteractionSerialEvaluatorDict', | |
| 'stop_value': 240, | |
| 'n_episode': 8 | |
| } | |
| }, | |
| 'other': { | |
| 'replay_buffer': { | |
| 'replay_buffer_size': 100000 | |
| } | |
| }, | |
| 'on_policy': False, | |
| 'cuda': True, | |
| 'multi_gpu': False, | |
| 'bp_update_sync': True, | |
| 'traj_len_inf': False, | |
| 'type': 'td3', | |
| 'priority': False, | |
| 'priority_IS_weight': False, | |
| 'random_collect_size': 10000, | |
| 'transition_with_policy_data': False, | |
| 'action_space': 'continuous', | |
| 'reward_batch_norm': False, | |
| 'multi_agent': False, | |
| 'cfg_type': 'TD3PolicyDict' | |
| }, | |
| 'exp_name': 'LunarLander-v2-TD3', | |
| 'seed': 0, | |
| 'wandb_logger': { | |
| 'gradient_logger': True, | |
| 'video_logger': True, | |
| 'plot_logger': True, | |
| 'action_logger': True, | |
| 'return_logger': False | |
| } | |
| } | |