to_numeric(errors='coerce') broken when using unhashable objects

[RogerThomas]

When using pandas.to_numeric with errors='coerce' and non hashable objects in the series/dataframe pandas returns the same series/dataframe as passed in, instead of coercing the non-numeric objects to nans.

python -c "import pandas as pd; s = pd.Series([[13], 1.0, 'apple']); print(pd.to_numeric(s, errors='coerce'))"

The error is due to this line

https://github.com/pydata/pandas/blob/4de83d25d751d8ca102867b2d46a5547c01d7248/pandas/src/inference.pyx#L572

More generally this will affect any call to maybe_convert_numeric that has unhashable objects in the values param.

Suggested fix might be to update this check with something like

if val.__hash__ is not None and val in na_values:
    blah

As there should never be a case where you would expect an unhashable object to be in a set of na_values

output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 2.7.11.final.0
python-bits: 64
OS: Linux
OS-release: 4.4.0-22-generic
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: None
LANG: en_IE.UTF-8

pandas: 0.18.1
nose: 1.3.7
pip: 8.1.2
setuptools: 19.0
Cython: None
numpy: 1.11.0
scipy: None
statsmodels: None
xarray: None
IPython: None
sphinx: None
patsy: None
dateutil: 2.5.3
pytz: 2016.4
blosc: None
bottleneck: None
tables: None
numexpr: None
matplotlib: None
openpyxl: None
xlrd: None
xlwt: None
xlsxwriter: None
lxml: None
bs4: None
html5lib: None
httplib2: None
apiclient: None
sqlalchemy: None
pymysql: None
psycopg2: None
jinja2: 2.8
boto: None
pandas_datareader: None

[jreback]

yeah, that's prob reasonable.