CODEWITHSUNDEEP
pythonpandaspandas-groupbyaggregation
Bijan
Bijan

Reputation: 8586

Pandas Groupby with no aggregation for single result

I have a Dataframe in Pandas that shows percentage of men in a city/state. The dataframe df looks like the following (note this is not my actual usage/data but my datatypes are similar)

STATE      CITY        PERC_MEN
ALABAMA    ABBEVILLE   41.3%
ALABAMA    ADAMSVILLE  53.5%
....
WYOMING    WRIGHT      46.6%

Each State/percentage of men combo will have exactly 1 value returned.

How do I show the city/population values for a given state? My code looks like the following (I need the first line where I groupby STATE because I do other stuff with the data)

for state, state_df in df.groupby(by=['STATE']):
    print(state_df.groupby(by=['CITY'])['PERC_MEN'])

However this prints <pandas.core.groupby.generic.SeriesGroupBy object at 0xXXXXXXX>

Normally for groupby's I use an aggregate like mean() or sum() but is there a way to just return the value?

Upvotes: 1

Views: 186

Answers (1)

wwnde
wwnde

Reputation: 26676

I wouldnt iterate dataframe.

Set index and slice

df=df.set_index(['STATE','CITY'])


df.xs(('ALABAMA', 'ABBEVILLE'), level=['STATE','CITY'])

or

df.loc[('ALABAMA', 'ABBEVILLE'),:]

Upvotes: 1

Related Questions