"ValueError: Dimension Mismatch" in Step 4: Run the ICLR algorithms

[Cozokim]

Hi !
When I'm launching te step 4 of the Part I. [ICRL in Virtual Environment] as the tutorial suggest, (https://github.com/Guiliang/ICRL-benchmarks-public/blob/main/virtual_env_tutorial.md)

Run GAIL
python train_gail.py ../config/mujuco_BlockedHalfCheetah/train_GAIL_HCWithPos-v0.yaml -n 5 -s 123

I'm encountering the following error when trying to run my code:

Traceback (most recent call last):
  File "/path/to/project/interface/train_gail.py", line 363, in <module>
    train(args)
  File "/path/to/project/interface/train_gail.py", line 290, in train
    model.learn(total_timesteps=int(config['PPO']['timesteps']),
  File "/path/to/project/stable_baselines3/ppo/ppo.py", line 255, in learn
    return super(PPO, self).learn(
  File "/path/to/project/stable_baselines3/common/on_policy_algorithm.py", line 231, in learn
    continue_training = self.collect_rollouts(self.env, callback, self.rollout_buffer, n_rollout_steps=self.n_steps)
  File "/path/to/project/stable_baselines3/common/on_policy_algorithm.py", line 198, in collect_rollouts
    callback.on_rollout_end()
  File "/path/to/project/stable_baselines3/common/callbacks.py", line 100, in on_rollout_end
    self._on_rollout_end()
  File "/path/to/project/stable_baselines3/common/callbacks.py", line 200, in _on_rollout_end
    callback.on_rollout_end()
  File "/path/to/project/stable_baselines3/common/callbacks.py", line 100, in on_rollout_end
    self._on_rollout_end()
  File "/path/to/project/constraint_models/constraint_net/gail_net.py", line 657, in _on_rollout_end
    metrics = self.discriminator.train(1, obs, acs)
  File "/path/to/project/constraint_models/constraint_net/gail_net.py", line 199, in train
    nominal_data, _ = self.prepare_nominal_data(nominal_obs, nominal_acs)
  File "/path/to/project/constraint_models/constraint_net/gail_net.py", line 352, in prepare_nominal_data
    concat = self.select_appropriate_dims(np.concatenate([obs, acs], axis=-1))
  File "<__array_function__ internals>", line 5, in concatenate
ValueError: all the input arrays must have the same number of dimensions, but the array at index 0 has 2 dimension(s) and the array at index 1 has 3 dimension(s)

The issue seems to arise from a mismatch in the dimensions of the input arrays when concatenating observations (obs) and actions (acs).

I have gym==0.21.0

I'm trying to run the ICLR algorithms using the Blocked Half-Cheetah environment to reproduce the results from Section 4.2 of the paper. Any help in resolving this would be greatly appreciated!

Thanks in Advance