Create a new column based on the multiple column values-CodePudding

I am sincerely sorry, if this is duplicated. I have searched a long time and still get error: "TypeError: _select_dispatcher() got an unexpected keyword argument 'na'" or "TypeError: invalid entry 0 in condlist: should be boolean ndarray".

I have a dataframe:

data_1 = {'A': ['Emo/3', 'Emo/4', 'Emo/1','Emo/3', '','Emo/3', 'Emo/4', 'Emo/1','Emo/3', '', 'Neu/5', 'Neu/2','Neu/5', 'Neu/2'],
        'Pos': ["repeat3", "repeat3", "repeat3", "repeat3", '',"repeat1", "repeat1", "repeat1", "repeat1", '', "repeat2", "repeat2","repeat2", "repeat2"],
        'B': [0, 0, 0, 0, '', 1, 2, 3, 4, '', 4, 2, 3, 1],'C': [0, 2, 1, 3, '', 4, 2, 3, 1, '', 4, 2, 3, 1]}
df_1 = pd.DataFrame(data_1)
df_1

    A   Pos B   C
0   Emo/3   repeat3 0   0
1   Emo/4   repeat3 0   2
2   Emo/1   repeat3 0   1
3   Emo/3   repeat3 0   3
4               
5   Emo/3   repeat1 1   4
6   Emo/4   repeat1 2   2
7   Emo/1   repeat1 3   3
8   Emo/3   repeat1 4   1
9               
10  Neu/5   repeat2 4   4
11  Neu/2   repeat2 2   2
12  Neu/5   repeat2 3   3
13  Neu/2   repeat2 1   1

I want to create a column D based on column B and C. If the satified with the criteria, put a number, if not leave as empty. Here is my code:

conditions = [
    
    df_1.loc[(df_1['B']==1)&(df_1['C']==1)],
    df_1.loc[(df_1['B']==2)&(df_1['C']==1)],
    df_1.loc[(df_1['B']==3)&(df_1['C']==1)],

]
choices = [1,1,0]
df_1['D'] = np.select(conditions, choices, default='')

Thanks!

CodePudding user response：

You shouldn't be using .loc in your conditions. Also, it's not a good idea to mix strings and numbers in a column, so you should set your default value to NaN instead of ''.

Try:

conditions = [(df_1['B']==1)&(df_1['C']==1),
              (df_1['B']==2)&(df_1['C']==1),
              (df_1['B']==3)&(df_1['C']==1)]
choices = [1,1,0]
df_1['D'] = np.select(conditions, choices, default=np.nan)

>>> df_1

        A      Pos  B  C    D
0   Emo/3  repeat3  0  0  NaN
1   Emo/4  repeat3  0  2  NaN
2   Emo/1  repeat3  0  1  NaN
3   Emo/3  repeat3  0  3  NaN
4                         NaN
5   Emo/3  repeat1  1  4  NaN
6   Emo/4  repeat1  2  2  NaN
7   Emo/1  repeat1  3  3  NaN
8   Emo/3  repeat1  4  1  NaN
9                         NaN
10  Neu/5  repeat2  4  4  NaN
11  Neu/2  repeat2  2  2  NaN
12  Neu/5  repeat2  3  3  NaN
13  Neu/2  repeat2  1  1  1.0