Cannot run PointMazeLeft-v0

[sff1019]

I've tried to run the codes with the following command:
python Train.py --seed 1 --env_name PointMazeLeft-v0 --policy_name SAC

However, I get a runtime error as follows:

Traceback (most recent call last):
  File "Train.py", line 523, in <module>
    main()
  File "Train.py", line 314, in main
    expert_state, expert_action)
  File "~/Off_Policy_Adversarial_Inverse_Reinforcement_Learning/Policies/Policy.py", line 201, in compute_pdf
    lprob = dist.log_prob(action).sum(-1, keepdim=True)
  File "~/lib/python3.7/site-packages/torch/distributions/normal.py", line 76, in log_prob
    return -((value - self.loc) ** 2) / (2 * var) - log_scale - math.log(math.sqrt(2 * math.pi))
RuntimeError: The size of tensor a (0) must match the size of tensor b (2) at non-singleton dimension 1

Could you check if the expert trajectories have all data needed? If I am correct, the trajectories seem to be missing 'next_states'.

Thank you