Filter by either index level or column when I don't know whether my field is an index level or column

Question

I have a DataFrame with a number of index levels and a number of columns. I have a field that I know is either the name of an index level or the name of a column, but I don't know which. I wish to filter my DataFrame on this field. If the field were an index level, I would do df[df.index.get_level_values(field) == 0], and if it were a column I'd do df[df[field] == 0], but since I don't know which it is I'm stuck. Is there a way to let pandas figure out whether the field name is in the index or the columns and grab the Index/Series as appropriate?

I know that I could reset the index and then set it back, but this seems dumb and potentially problematic if not all of the index levels are named.

So catch the `KeyError` when it can't find it in the columns? That seems... ugly — BallpointBen, Jan 25 '19 at 20:02
https://stackoverflow.com/questions/12265451/ask-forgiveness-not-permission-explain — roganjosh, Jan 25 '19 at 20:03
Actually, I posted a link that wasn't the one I was thinking of :P It's still probably the easiest way to my mind. — roganjosh, Jan 25 '19 at 20:05

score 0 · Answer 1 · answered Jan 25 '19 at 21:02

0

Since you mentioned it's either column or index label, then try:

if field in list(df.columns):
   #do df[df[field] == 0]
else:
   #do df[df.index.get_level_values(field) == 0]

answered Jan 25 '19 at 21:02

Artem Trunov

1,340
9
16

Filter by either index level or column when I don't know whether my field is an index level or column

1 Answers1