Replace every row in dataframe with a list

Question

I have a dataframe with a column containing a list with values to replace each column, and I'm not sure how to move the list to do this. Here is an example dataframe:

            A    B    C    D
2020-07-31  0    0    0    [2,3,4]
2020-08-31  0    0    0    [5,6,7]
2020-09-30  0    0    0    [8,9,10]
2020-10-31  0    0    0    [0,1,2]

I would want to replace column A, B, and C with D like so:

            A    B    C    
2020-07-31  2    3    4  
2020-08-31  5    6    7 
2020-09-30  8    9    10 
2020-10-31  0    1    2

What is the most efficient way to do this?

EDIT: column D is in the original dataframe, I want to move the values from D into A, B, and C and then drop the column D

Please refer to this already answered question: https://stackoverflow.com/questions/35491274/split-a-pandas-column-of-lists-into-multiple-columns — dp6000, Jul 21 '22 at 20:16

score 2 · Accepted Answer · edited Jul 21 '22 at 20:36

2

You can use:

df[:] = df.pop('D').to_list()
# or specifying columns
# df[['A', 'B', 'C']] = df.pop('D').tolist()

print(df)

# Output
            A  B   C
2020-07-31  2  3   4
2020-08-31  5  6   7
2020-09-30  8  9  10
2020-10-31  0  1   2

edited Jul 21 '22 at 20:36

mozway

194,879
13
39
75

answered Jul 21 '22 at 20:06

Corralien

109,409
8
28
52

I read it as they wanted to go from A,B,C to -> D not the other way around – Jordan Hyatt Jul 21 '22 at 20:07
1

Are you really sure? :) *I would want to replace column A, B, and C with D like so:* – Corralien Jul 21 '22 at 20:10
I edited my question to clarify – Roxanne Jul 21 '22 at 20:15
@Roxanne. I updated my answer according your comment. Can you check it please? – Corralien Jul 21 '22 at 20:18
@Corralien yes this works, but is there a way to do this without specifying columns 'A', 'B', and 'C'? In my actual dataset I have a few hundred columns so I wouldn't want to write them all down. – Roxanne Jul 21 '22 at 20:26
Like `df = df.pop('D').tolist()`? Do you have some columns after D? – Corralien Jul 21 '22 at 20:28
1

use `df[:] = df.pop('D').to_list()`, `:` is evaluated after the `pop` occurred ;) – mozway Jul 21 '22 at 20:37

Jordan Hyatt · Answer 2 · 2022-07-21T20:05:38.883

0

Assuming col "D" is not already in the original dataset and those dates are the index

df['D'] = df.apply(lambda ser: list(ser), axis=1)

If you only want to keep the index and column D

df = df[['D']]

edited Jul 21 '22 at 20:05

answered Jul 21 '22 at 20:00

Jordan Hyatt

435
3
7

score 0 · Answer 3 · answered Jul 21 '22 at 20:08

0

df[['A', 'B', 'C']] = df.D.tolist()

The proposed solution assumes that Column D will always have a list with exact three items.

answered Jul 21 '22 at 20:08

dp6000

473
5
15

Replace every row in dataframe with a list

3 Answers3