python - TypeError: ("sort_values() got multiple values for argument 'axis'", 'occurred at index SUMLEV')

Question

Why this code

cdf = census_df[census_df['SUMLEV'] == 50]
cdf = cdf.apply(lambda x:x.sort_values('CENSUS2010POP', axis=0, ascending=False)).reset_index(drop=True)
cdf = cdf.groupby('STNAME').head(3)
cdf.head(20)

gives the following error

TypeError: ("sort_values() got multiple values for argument 'axis'", 'occurred at index SUMLEV')

While this code works fine

cdf = census_df[census_df['SUMLEV'] == 50]
cdf = cdf.groupby('STNAME')
cdf = cdf.apply(lambda x:x.sort_values('CENSUS2010POP', axis=0, ascending=False)).reset_index(drop=True)
cdf = cdf.groupby('STNAME').head(3)
cdf.head(20)

But here I needed to do twice groupby, first before sorting and after sorting to pick top 3 values. I wanted first to sort, then group, then pick 3 for every group.

csv file can be found here

Thanks!

Answer 1

You need to change the line:

cdf = cdf.apply(lambda x:x.sort_values('CENSUS2010POP', axis=0, ascending=False)).reset_index(drop=True)

to:

cdf = cdf.sort_values('CENSUS2010POP', ascending=False).reset_index(drop=True)

When .sort_values() is applied to data frame, it will sort all columns by the column you specified, i.e. CENSUS2010POP in this case. You don't have to apply the sort to all columns.