TypeError: clip()

[Exc1tedFr0g]

遇到错误 Traceback (most recent call last): File "wheel_legged_gym/scripts/train.py", line 55, in train(args) File "wheel_legged_gym/scripts/train.py", line 44, in train env=env, name=args.task, args=args File "/home/excitedfrog/Wheel-Legged-Gym/wheel_legged_gym/utils/task_registry.py", line 218, in make_alg_runner env, train_cfg_dict, self.log_dir, device=args.rl_device File "/home/excitedfrog/Wheel-Legged-Gym/wheel_legged_gym/rsl_rl/runners/on_policyrunner.py", line 89, in init , _ = self.env.reset() File "/home/excitedfrog/Wheel-Legged-Gym/wheel_legged_gym/envs/base/base_task.py", line 146, in reset self.num_envs, self.num_actions, device=self.device, requires_grad=False File "/home/excitedfrog/Wheel-Legged-Gym/wheel_legged_gym/envs/wheel_legged_vmc/wheel_legged_vmc.py", line 93, in step self.action_fifo[torch.arange(self.num_envs), self.action_delay_idx, :] File "/home/excitedfrog/Wheel-Legged-Gym/wheel_legged_gym/envs/wheel_legged_vmc/wheel_legged_vmc.py", line 415, in _compute_torques torques * self.torques_scale, -self.torque_limits, self.torque_limits TypeError: clip(): argument 'min' (position 2) must be Number, not Tensor 是否是环境问题？ torch版本1.8.1 torchvision版本0.9.1