Can't run deeplearning example due to AttributeError

[no-clue-what-to-do]

Contribution guidelines

• I've read the contribution guidelines.
• The documentation does not mention anything about my problem.
• There are no open or closed issues that are related to my problem.

Description

Wehn trying to build the AtomClassifier, I get an attribute error because some string is empty and can't be split

Expected behaviour

No response

Actual behaviour

When I run the deep learning example on the atom page, I get the following error message

  File "C:\Users\i13500020\.spyder-py3\IST\untitled0.py", line 74, in <module>
    atom = ATOMClassifier(*data, n_rows=0.1, n_jobs=6, warnings=False, verbose=2)

  File "C:\Users\i13500020\AppData\Roaming\Python\Python38\site-packages\typeguard\__init__.py", line 1033, in wrapper
    retval = func(*args, **kwargs)

  File "C:\Users\i13500020\AppData\Roaming\Python\Python38\site-packages\atom\api.py", line 316, in __init__
    super().__init__(

  File "C:\Users\i13500020\AppData\Roaming\Python\Python38\site-packages\atom\basetransformer.py", line 63, in __init__
    setattr(self, key, value)

  File "C:\Users\i13500020\AppData\Roaming\Python\Python38\site-packages\atom\basepredictor.py", line 58, in __setattr__
    super().__setattr__(item, value)

  File "C:\Users\i13500020\AppData\Roaming\Python\Python38\site-packages\atom\basetransformer.py", line 139, in experiment
    mlflow.sklearn.autolog(disable=True)

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\mlflow\utils\autologging_utils\__init__.py", line 468, in autolog
    return _autolog(*args, **kwargs)

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\mlflow\sklearn\__init__.py", line 943, in autolog
    _, estimators_to_patch = zip(*_all_estimators())

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\mlflow\sklearn\utils.py", line 728, in _all_estimators
    return all_estimators()

  File "C:\Users\i13500020\AppData\Roaming\Python\Python38\site-packages\sklearn\utils\__init__.py", line 1174, in all_estimators
    from ._testing import ignore_warnings

  File "C:\Users\i13500020\AppData\Roaming\Python\Python38\site-packages\sklearn\utils\_testing.py", line 531, in <module>
    _in_unstable_openblas_configuration(),

  File "C:\Users\i13500020\AppData\Roaming\Python\Python38\site-packages\sklearn\utils\__init__.py", line 93, in _in_unstable_openblas_configuration
    modules_info = threadpool_info()

  File "C:\Users\i13500020\AppData\Roaming\Python\Python38\site-packages\sklearn\utils\fixes.py", line 162, in threadpool_info
    return threadpoolctl.threadpool_info()

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\threadpoolctl.py", line 124, in threadpool_info
    return _ThreadpoolInfo(user_api=_ALL_USER_APIS).todicts()

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\threadpoolctl.py", line 340, in __init__
    self._load_modules()

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\threadpoolctl.py", line 373, in _load_modules
    self._find_modules_with_enum_process_module_ex()

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\threadpoolctl.py", line 485, in _find_modules_with_enum_process_module_ex
    self._make_module_from_path(filepath)

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\threadpoolctl.py", line 515, in _make_module_from_path
    module = module_class(filepath, prefix, user_api, internal_api)

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\threadpoolctl.py", line 606, in __init__
    self.version = self.get_version()

  File "C:\Users\i13500020\Anaconda3\lib\site-packages\threadpoolctl.py", line 646, in get_version
    config = get_config().split()

AttributeError: 'NoneType' object has no attribute 'split'

I get this error message every time I try to build a regressor or classfier. I had a bunch of issues installing atom-ml, so maybe that's where this comes from? Any guidance would be very welcome

Steps to reproduce

No response

Python and package version

• Python: 3.8.5
• ATOM: 4.13.1

[tvdboom]

Hi, I can't reproduce the error and, to be honest, the error doesn't seem to have anything to do with the data used nor ATOM itself. Couple of suggestions:

• What were the exact problems you encountered during installation? Since the problem arises in mlflow.sklearn.autolog(disable=True), reinstalling the mlflow library could help.
• Try installing from the development branch: pip install git+https://github.com/tvdboom/ATOM.git@development. This branch has some updates that could potentially solve this issue.

Let me know if any of his worked.

[no-clue-what-to-do]

Thanks! Yea, after looking into it more yesterday I think that mlflow is the issue. Really weird since I haven't had any trouble with it before

I managed to succesfully install ATOM after restarting some things and I didn't document the error. I think it was a path issue though. If it comes up again when installing the development branch and it doesn't look like its a problem with my setup I'll share the error