Is it possible to pass a column name as a parameter to a method and if so, how can I do that?

Question

I want to replace values in different columns. Namely, if they contain a "Y", then with the value "Y" and if they contain an "N", then with the value "N".

Since I do this for many columns, the code looks very redundant and I was wondering if I can outsource the always similar procedure (only the column name changes) to a method.

Unfortunately I have no idea how to pass the column name as parameter to the method I write then. Is this possible? I am a Python beginner and have never seen this before!

Thanks a lot for your help!

df = pd.DataFrame({'A': ['J', 'Ja', 'N', 'N  n'],
                   'B': ['Nö', 'Ja', 'N', 'N  n'],
                   'C': ['Jup', 'Ja', 'N', 'N  n']})

df.loc[df['A'].str.contains('J'), 'A'] = 'J'
df.loc[df['A'].str.contains('N'), 'A'] = 'N'

df.loc[df['B'].str.contains('J'), 'B'] = 'J'
df.loc[df['B'].str.contains('N'), 'B'] = 'N'

df.loc[df['C'].str.contains('J'), 'C'] = 'J'
df.loc[df['C'].str.contains('N'), 'C'] = 'N'

def clear_values(column_namestr):
    column_name = [column_namestr]
    print(type(column_name))
    #df.loc[df[column_name].str.contains('X'), 'spaltennamen'] = 'X'
    #df.loc[df[column_name].str.contains('o'), 'spaltennamen'] = 'o'

column_namestr = 'a'
df2 = clear_values(column_namestr)

But does not work...

Asked By: user20880339

||

Source

Answer 1

If you want to replicate

df.loc[df['C'].str.contains('J'), 'C'] = 'J'
df.loc[df['C'].str.contains('N'), 'C'] = 'N'

with a function you can do the following

def clear_values(name1, name2):
    df.loc[df[name1].str.contains(name2), name1] = name2
clear_values('C', 'J')
clear_values('C', 'N')

If you only want to change the column you can do this:

def clear_values(column):
    df.loc[df[column].str.contains('J'), column] = 'J'
    df.loc[df[column].str.contains('N'), column] = 'N'
clear_values('B')
clear_values('C')

Answered By: White Wizard

Is it possible to pass a column name as a parameter to a method and if so, how can I do that?

Question:

Answers: