{ "T": 5, "action_dim": 5, "algo": "ql", "batch_size": 256, "beta_schedule": "vp", "device": "cuda:4", "dir": "dql/results", "discount": 0.99, "early_stop": false, "env_name": "cube-double-play-v0", "eta": 1.0, "eval_episodes": 100, "eval_freq": 100, "exp": "exp", "gn": 7.0, "goal_dim": 3, "lcb_coef": 4.0, "lr": 0.0003, "lr_decay": true, "max_action": 1.0, "max_q_backup": false, "ms": "offline", "num_epochs": 2000, "num_steps_per_epoch": 1000, "output_dir": "dql/results", "p": 0.2, "replan": 15, "reward_tune": "cql_antmaze", "save_best_model": false, "seed": 0, "state_dim": 37, "target_steps": 10, "tau": 0.005, "top_k": 1 }