Replace values of a column using assign method-CodePudding

I have below data frame

import pandas as pd, numpy as np
df = pd.DataFrame(np.random.randint(0,100,size=(100, 2)), columns=list('AB'))
>>> df
     A   B
0   50  59
1    7  15
2   35  54
3   86  62
4   75  68
..  ..  ..
95  16  32
96   2  83
97  65  18
98  91  39
99  33  50

Now let say I want to replace values of the first column using assign method. So I can write

df.assign(A = lambda x : x['A'] * 2)

But instead of calling column by name, I want to use index, i.e. something like

df.assign(:0 = lambda x : x.iloc[:, 0] * 2)

Above code is generating error. Is there any way to achieve this?

CodePudding user response：

You can store the index of your column you want to update (assuming i as the index and 0 as the location), and use assign as follows:

i = 0
colname = df.columns[i]
res = df.assign(**{colname : lambda x : x[colname] * 2})

prints:

>>> print(df)
     A   B
0   38  32
1   19  17
2   38  69
3   60   4
4   26  81
..  ..  ..
95  46  46
96  87  44
97  43  42
98  18  77
99  25  74

[100 rows x 2 columns]

>>> print(res)
      A   B
0    76  32
1    38  17
2    76  69
3   120   4
4    52  81
..  ...  ..
95   92  46
96  174  44
97   86  42
98   36  77
99   50  74

[100 rows x 2 columns]

CodePudding user response：

Why don't you use the following:

df.iloc[:,1] = df.iloc[:,1] * 2

this multiplies all the values in the 2nd column by 2, as requested