Pandas Groupby - Apply method doesn't handle arguments correctly

The apply method on a groupby object appears to send all arguments to the applied function. As in, if you try to do df.groupby(level='dt').apply(func, axis=1), the axis=1 argument is passed to func rather than being parsed by apply as intended. The behavior is correct, however, in the DataFrame.apply method.

Comment From: jreback

See docs

you are probably looking to do:

df.groupby(level='dt',axis=1).apply(func)