BUG: MultiIndex .loc fails with all numpy arrays

[innominate227]

Code Sample, a copy-pastable example if possible

import pandas as pd
import numpy as np

a = [10,20,30]
b = [1,2,3]
index = pd.MultiIndex.from_product([a,b])
df = pd.DataFrame(np.arange(len(index)), index=index, columns=['Data'])

slice1 = df.loc[([10,20],            [2,3]),           'Data']
slice2 = df.loc[(np.array([10,20]),  [2,3]),           'Data']
slice3 = df.loc[([10,20],            np.array([2,3])), 'Data']
slice4 = df.loc[(np.array([10,20]),  np.array([2,3])), 'Data']

Problem description

slice4 in the sample fails with TypeError: '(array([10, 20]), array([2, 3]))' is an invalid key.
slice1-3 all work as expected.

Expected Output

expect slicing to work with numpy arrays the same as lists.

Output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 3.5.2.final.0
python-bits: 64
OS: Windows
OS-release: 10
machine: AMD64
processor: Intel64 Family 6 Model 45 Stepping 7, GenuineIntel
byteorder: little
LC_ALL: None
LANG: None
LOCALE: None.None

pandas: 0.22.0
pytest: 2.9.2
pip: 8.1.2
setuptools: 27.2.0
Cython: 0.24.1
numpy: 1.14.0
scipy: 1.0.0
pyarrow: None
xarray: None
IPython: 5.1.0
sphinx: 1.4.6
patsy: 0.4.1
dateutil: 2.5.3
pytz: 2016.6.1
blosc: None
bottleneck: 1.2.1
tables: 3.4.2
numexpr: 2.6.4
feather: None
matplotlib: 2.1.2
openpyxl: 2.3.2
xlrd: None
xlwt: 1.1.2
xlsxwriter: 0.9.3
lxml: 3.6.4
bs4: 4.5.1
html5lib: 0.9999999
sqlalchemy: 1.0.13
pymysql: None
psycopg2: None
jinja2: 2.8
s3fs: 0.1.2
fastparquet: None
pandas_gbq: None
pandas_datareader: None

[jorisvandenbossche]

@innominate227 Thanks for the report! That indeed looks like a bug (I at least cannot think of a reason why we would do it like this)

[spacesphere]

I'd like to work on it if nobody minds. May I?

[jorisvandenbossche]

Go ahead! If you have any question, feel free to ask here, or on the gitter channel (https://gitter.im/pydata/pandas)
(note: it's a bit hard to predict without looking into details at it, but it might not be the most easy issue)