[BUG] ColumnTransformer error when you pass pandas.Index as columns

[codeananda]

Minimal code to reproduce:

import numpy as np
import cudf
from cuml.experimental.preprocessing import ColumnTransformer
from cuml.preprocessing import RobustScaler

data = np.arange(12).reshape(3, 4)
X = cudf.DataFrame(data=data, columns=['a', 'b', 'c', 'd'])

col_index= X.columns[:2]
col_index_list = list(col_index)

ct = ColumnTransformer([
    ('scaler', RobustScaler(), col_index)
])

ct.fit_transform(X)

Note: swap col_index for col_index_list and it works fine.

Error message

---------------------------------------------------------------------------
Empty                                     Traceback (most recent call last)
/opt/conda/envs/rapids/lib/python3.7/site-packages/joblib/parallel.py in dispatch_one_batch(self, iterator)
    819             try:
--> 820                 tasks = self._ready_batches.get(block=False)
    821             except queue.Empty:

/opt/conda/envs/rapids/lib/python3.7/queue.py in get(self, block, timeout)
    166                 if not self._qsize():
--> 167                     raise Empty
    168             elif timeout is None:

Empty: 

During handling of the above exception, another exception occurred:

ValueError                                Traceback (most recent call last)
<ipython-input-61-c058d4fbb573> in <module>
      9 ])
     10 
---> 11 ct.fit_transform(X)

/opt/conda/envs/rapids/lib/python3.7/site-packages/cuml/internals/api_decorators.py in inner_get(*args, **kwargs)
    584 
    585                 # Call the function
--> 586                 ret_val = func(*args, **kwargs)
    587 
    588             return cm.process_return(ret_val)

/opt/conda/envs/rapids/lib/python3.7/site-packages/cuml/_thirdparty/sklearn/preprocessing/_column_transformer.py in fit_transform(self, X, y)
    876         self._validate_remainder(X)
    877 
--> 878         result = self._fit_transform(X, y, _fit_transform_one)
    879 
    880         if not result:

/opt/conda/envs/rapids/lib/python3.7/site-packages/cuml/_thirdparty/sklearn/preprocessing/_column_transformer.py in _fit_transform(self, X, y, func, fitted)
    813                     message=self._log_message(name, idx, len(transformers)))
    814                 for idx, (name, trans, column, weight) in enumerate(
--> 815                         self._iter(fitted=fitted, replace_strings=True), 1))
    816         except ValueError as e:
    817             if "Expected 2D array, got 1D array instead" in str(e):

/opt/conda/envs/rapids/lib/python3.7/site-packages/joblib/parallel.py in __call__(self, iterable)
   1039             # remaining jobs.
   1040             self._iterating = False
-> 1041             if self.dispatch_one_batch(iterator):
   1042                 self._iterating = self._original_iterator is not None
   1043 

/opt/conda/envs/rapids/lib/python3.7/site-packages/joblib/parallel.py in dispatch_one_batch(self, iterator)
    829                 big_batch_size = batch_size * n_jobs
    830 
--> 831                 islice = list(itertools.islice(iterator, big_batch_size))
    832                 if len(islice) == 0:
    833                     return False

/opt/conda/envs/rapids/lib/python3.7/site-packages/cuml/_thirdparty/sklearn/preprocessing/_column_transformer.py in <genexpr>(.0)
    812                     message_clsname='ColumnTransformer',
    813                     message=self._log_message(name, idx, len(transformers)))
--> 814                 for idx, (name, trans, column, weight) in enumerate(
    815                         self._iter(fitted=fitted, replace_strings=True), 1))
    816         except ValueError as e:

/opt/conda/envs/rapids/lib/python3.7/site-packages/cuml/_thirdparty/sklearn/preprocessing/_column_transformer.py in _safe_indexing(X, indices, axis)
    238 
    239     if hasattr(X, "iloc"):
--> 240         return _pandas_indexing(X, indices, indices_dtype, axis=axis)
    241     elif hasattr(X, "shape"):
    242         return _array_indexing(X, indices, indices_dtype, axis=axis)

/opt/conda/envs/rapids/lib/python3.7/site-packages/cuml/_thirdparty/sklearn/preprocessing/_column_transformer.py in _pandas_indexing(X, key, key_dtype, axis)
    263         # Work-around for indexing with read-only key in pandas
    264         # FIXME: solved in pandas 0.25
--> 265         key = np.asarray(key)
    266         key = key if key.flags.writeable else key.copy()
    267     elif isinstance(key, tuple):

/opt/conda/envs/rapids/lib/python3.7/site-packages/cupy/_creation/from_data.py in asarray(a, dtype, order)
     64 
     65     """
---> 66     return _core.array(a, dtype, False, order)
     67 
     68 

cupy/_core/core.pyx in cupy._core.core.array()

cupy/_core/core.pyx in cupy._core.core.array()

cupy/_core/core.pyx in cupy._core.core._send_object_to_gpu()

ValueError: Unsupported dtype object

Using cuml 21.08.02 installed on Paperspace Gradient via this container:
rapidsai/rapidsai-core:21.08-cuda11.0-runtime-ubuntu18.04-py3.7

[github-actions]

This issue has been labeled inactive-30d due to no recent activity in the past 30 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed. This issue will be labeled inactive-90d if there is no activity in the next 60 days.

[viclafargue]

Thanks for opening the issue. #4481 should solve it.