[Python] Picking N largest from a group in Pandas

I want to get the top 5 from an aggregate mean measure.

cohort = ['year','month','maincategory','subcategory']
poorresult = stmpoor.groupby(cohort,as_index=False).agg({'resolvetime': ['count','mean'], 'grouptransfers': ['mean']}).apply(lambda x: x.nlargest(5, ('resolvetime','mean'))


but that threw an error mentioning the keep keyword. Error said

ValueError: keep must be either "first", "last" or "all"


I looked up the keep keyword and then I tried

poorresult = stmpoor.groupby(cohort,as_index=False).agg({'resolvetime': ['count','mean'], 'grouptransfers': ['mean']})# .apply(lambda x: x.nlargest(5, ('resolvetime','mean'),keep='all'))


but that did not work either. The error was

TypeError: Series.nlargest() got multiple values for argument 'keep'


What is the correct syntax?

Continue reading...