如何在不引用旧列名和不创建新数据框的情况下更改列名？

Question

I know I can change column names by doing:

df = df.rename(columns = {old_name: new_name})

But I have to rename a lot and the old names are long so I wanna just pass in a list for the new names. If I make the new columns its own data frame this works fine:

ranks = sw[sw.columns[9:15]]
ranks.columns = ['ranking_1', 'ranking_2', 'ranking_3', 'ranking_4', 'ranking_5', 'ranking_6']

However if I try to just do it with the slice of columns without changing it to its own df, nothing happens:

sw[sw.columns[9:15]].columns = ['ranking_1', 'ranking_2', 'ranking_3', 'ranking_4', 'ranking_5', 'ranking_6']

How can I rename the columns and keep them in the original data frame?

Answer 1

This:

sw[sw.columns[9:15]] = ['ranking_1', 'ranking_2', 'ranking_3', 'ranking_4', 'ranking_5', 'ranking_6']

changes the data of your sw , not the column names. A manual way to change the column names can be:

# new column names
new_cols = ['ranking_1', 'ranking_2', 'ranking_3', 'ranking_4', 'ranking_5', 'ranking_6']

# slice the columns name, replace the part you want and concatenate
sw.columns = list(sw.columns[:9]) + new_cols + list(sw.columns[15:])

Or you can also build the dictionary and use rename :

rename_dict = {a:b for a,b in zip(new_cols, sw.columns[9:15])}
sw.rename(columns=rename_dict)

Answer 2

Do you mean something like this?

import pandas as pd

df = pd.DataFrame(columns=['A','B','C','D','E',
                           'F','G','H','I','J',
                           'K','L','M','N','O'])
col = list(df.columns)                # get them all
rankcol = ['ranking_1', 'ranking_2', 'ranking_3', 
           'ranking_4', 'ranking_5', 'ranking_6']

df.columns = col[:9]+rankcol