在 Pandas DataFrame 中用 NaN 替换字符串（来自列表）

Question

I have a DataFrame df and a list of strings colnames

df = pd.DataFrame({'foo': ['a', 'foo', 'c' , 'bar'], 'bar': ['foo', 'b', 'c', 'bar']})

colnames = ['foo', 'bar']

I want to replace the words contained in colnames in my DataFrame df with 'NaN'. However, I only want to replace the values and not the column names. The output should look like:

    foo bar
0   a   NaN
1   NaN b
2   c   c
3   NaN NaN

Answer 1

Try this :

df.replace(colnames, np.nan)

Output :

   foo  bar
0    a  NaN
1  NaN    b
2    c    c
3  NaN  NaN

Answer 2

Here isin and mask :

df.mask(df.isin(colnames))

Output:

   foo  bar
0    a  NaN
1  NaN    b
2    c    c
3  NaN  NaN

Answer 3

Let us try replace

df.replace(df.columns,np.nan)
   foo  bar
0    a  NaN
1  NaN    b
2    c    c
3  NaN  NaN

Answer 4

You can also use an index mask:

df[~df.isin(colnames)] 

output

   foo  bar
0    a  NaN
1  NaN    b
2    c    c
3  NaN  NaN