vocab_sample path

[Darkdragon84]

In the config.yaml the part about the experiment vocab data requires vocab_sample to be the full path to a .vocab file:

experiments:

  # the vocabulary dataset to be tested, expected to be a single-column list of words
  # with .vocab extension. # Provide the full path to the .vocab file here, e.g.

  # vocab_sample: /path/to/test_sample.vocab

However, when doing this, the default_workflow pipeline crashes with

Traceback (most recent call last):
  File "/home/valentin/miniconda3/envs/ldt/lib/python3.9/runpy.py", line 197, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/home/valentin/miniconda3/envs/ldt/lib/python3.9/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/home/valentin/miniconda3/envs/ldt/lib/python3.9/site-packages/ldt/experiments/default_workflow/__main__.py", line 13, in <module>
    ldt.experiments.default_workflow()
  File "/home/valentin/miniconda3/envs/ldt/lib/python3.9/site-packages/ldt/experiments/default_workflow/default_workflow.py", line 37, in default_workflow
    neighborhoods = ldt.experiments.VectorNeighborhoods(
  File "/home/valentin/miniconda3/envs/ldt/lib/python3.9/site-packages/ldt/experiments/neighbors.py", line 82, in __init__
    super(VectorNeighborhoods, self).__init__(
  File "/home/valentin/miniconda3/envs/ldt/lib/python3.9/site-packages/ldt/experiments/metadata.py", line 93, in __init__
    self._load_dataset(dataset=dataset)
  File "/home/valentin/miniconda3/envs/ldt/lib/python3.9/site-packages/ldt/experiments/neighbors.py", line 124, in _load_dataset
    file = [x for x in os.listdir(dataset_path) if x.endswith(".vocab")][0]
NotADirectoryError: [Errno 20] Not a directory: '/mnt/Data/libs/ldt/Wiki201308/Wiki201308.vocab/'

What works is specifying the path to the folder containing the .vocab file, not the full path to the .vocab file. Please adjust either the comment in the config.yaml or the code (I suppose the former is easier :-) ). Thanks!