RuntimeError: stack expects a non-empty TensorList

[youying12123]

I'm trying to run projects/geometric_models/lane_occupancy/run.py on USA_US101-26_1_T-1.xml and am getting this error.

Traceback (most recent call last):
  File "/home/songze/Codes/crgeo-main/projects/geometric_models/lane_occupancy/run.py", line 21, in <module>
    main()
  File "/home/songze/anaconda3/envs/crgeo/lib/python3.8/site-packages/hydra/main.py", line 94, in decorated_main
    _run_hydra(
  File "/home/songze/anaconda3/envs/crgeo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 394, in _run_hydra
    _run_app(
  File "/home/songze/anaconda3/envs/crgeo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 457, in _run_app
    run_and_report(
  File "/home/songze/anaconda3/envs/crgeo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
    raise ex
  File "/home/songze/anaconda3/envs/crgeo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 220, in run_and_report
    return func()
  File "/home/songze/anaconda3/envs/crgeo/lib/python3.8/site-packages/hydra/_internal/utils.py", line 458, in <lambda>
    lambda: hydra.run(
  File "/home/songze/anaconda3/envs/crgeo/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 132, in run
    _ = ret.return_value
  File "/home/songze/anaconda3/envs/crgeo/lib/python3.8/site-packages/hydra/core/utils.py", line 260, in return_value
    raise self._return_value
  File "/home/songze/anaconda3/envs/crgeo/lib/python3.8/site-packages/hydra/core/utils.py", line 186, in run_job
    ret.return_value = task_function(task_cfg)
  File "/home/songze/Codes/crgeo-main/projects/geometric_models/lane_occupancy/run.py", line 17, in main
    project.run(cfg_obj.cmd)
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/base_project.py", line 60, in run
    self._commands[subcmd](self)
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/base_project.py", line 18, in _decorator
    profile(partial(func, self=self), args=args, kwargs=kwargs)
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/debugging/profiling.py", line 88, in profile
    raise catched_exception
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/debugging/profiling.py", line 32, in profile
    func(*args, **kwargs)
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/geometric/project/base_geometric_project.py", line 329, in train
    trainer.launch_trainer(
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/geometric/training/geometric_trainer.py", line 176, in launch_trainer
    self.load_dataset_and_setup_trainer(
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/geometric/training/geometric_trainer.py", line 310, in load_dataset_and_setup_trainer
    results = self.train_orchestrator(
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/geometric/training/geometric_trainer.py", line 395, in train_orchestrator
    self.train(None, model, experiment, train_loader, test_loader,
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/geometric/training/geometric_trainer.py", line 532, in train
    train_losses, train_info_dicts = self.train_loop(
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/geometric/training/geometric_trainer.py", line 626, in wrapper
    return func(*args, **kwargs)
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/geometric/training/geometric_trainer.py", line 682, in train_loop
    train_output, train_loss_step, info_dict = execute_step()
  File "/home/songze/Codes/crgeo-main/commonroad_geometric/learning/geometric/training/geometric_trainer.py", line 667, in execute_step
    batch = ctx.model.train_preprocess(batch)
  File "/home/songze/Codes/crgeo-main/projects/geometric_models/lane_occupancy/models/occupancy/occupancy_model.py", line 432, in train_preprocess
    rw_directed = torch.stack([row for row in rw if row.unique(return_counts=True)[1].max().item() == 1 ])
RuntimeError: stack expects a non-empty TensorList