Description
Code Sample, a copy-pastable example if possible
It's possible to create a Categorical of mixed dtypes, with at least one tuple, if the first element is a tuple.
s = pd.Categorical([('a', 'a'), ('a', 'b'), ('b', 'a'), 'c'])
s
Out[21]:
[(a, a), (a, b), (b, a), c]
Categories (4, object): [(a, a), (a, b), (b, a), c]Does not work if first element is not a tuple.
s = pd.Categorical(['c', ('a', 'b'), ('b', 'a'), 'c'])
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-22-0f4b5f338532> in <module>()
----> 1 s = pd.Categorical(['c', ('a', 'b'), ('b', 'a'), 'c'])
~/Documents/siv-dev/projects/open-source/pandas/pandas/core/arrays/categorical.py in __init__(self, values, categories, ordered, dtype, fastpath)
328 # _sanitize_array coerces np.nan to a string under certain versions
329 # of numpy
--> 330 values = maybe_infer_to_datetimelike(values, convert_dates=True)
331 if not isinstance(values, np.ndarray):
332 values = _convert_to_list_like(values)
~/Documents/siv-dev/projects/open-source/pandas/pandas/core/dtypes/cast.py in maybe_infer_to_datetimelike(value, convert_dates)
893 if not is_list_like(v):
894 v = [v]
--> 895 v = np.array(v, copy=False)
896
897 # we only care about object dtypes
ValueError: setting an array element with a sequenceProblem description
While writing tests for the #20439 fix, I found that it's not possible create Categorical objects with mixed dtypes, with at least one tuple, if the first item is not a tuple.
Expected Output
Constructor should accept non-tuple first arguments and return a Categorical
Output of pd.show_versions()
pandas: 0.23.0.dev0+970.g415012f
pytest: 3.1.2
pip: 10.0.1
setuptools: 36.0.1
Cython: 0.27.3
numpy: 1.14.3
scipy: None
pyarrow: None
xarray: None
IPython: 6.1.0
sphinx: 1.6.2
patsy: None
dateutil: 2.6.0
pytz: 2017.2
blosc: None
bottleneck: None
tables: None
numexpr: None
feather: None
matplotlib: 2.0.2
openpyxl: 2.4.8
xlrd: 1.0.0
xlwt: 1.2.0
xlsxwriter: 0.9.6
lxml: None
bs4: None
html5lib: 0.999999999
sqlalchemy: None
pymysql: None
psycopg2: None
jinja2: 2.9.6
s3fs: 0.1.0
fastparquet: None
pandas_gbq: None
pandas_datareader: None