Convert pandas df columns to paired list

Question

If I have a df like this:

df1 = pd.DataFrame({'col1': [1, 1, 2, 2],
                   'col2': [10, 20, 10, 20]})

How would I get a list that pairs each row of col1 and col2 like so:

outputlist = [[1, 10], [1, 20], [2, 10], [2, 20]]

I've found ways to turn lists into df columns, but not the other way around!

score 2 · Accepted Answer · answered Jan 03 '19 at 19:51

2

You could do (see tolist):

import pandas as pd

df1 = pd.DataFrame({'col1': [1, 1, 2, 2],
                   'col2': [10, 20, 10, 20]})

result = df1.values.tolist()
print(result)

Output

[[1, 10], [1, 20], [2, 10], [2, 20]]

answered Jan 03 '19 at 19:51

Dani Mesejo

61,499
6
49
76

Not even sure `tolist` is needed. – kabanus Jan 03 '19 at 19:52
@kabanus What do you mean? – Dani Mesejo Jan 03 '19 at 19:53
From OP's requirements the value array might be good enough (maybe better?) Anyway, nice answer. – kabanus Jan 03 '19 at 19:54
How would I select specific columns? eg if I have a `col3`, but I only want `col1` and `col2` to be in the output? I realize I could just drop `col3` and then use `.tolist()`, but I was just curious. – Liquidity Jan 03 '19 at 19:54
1

I guess you could do: `df1[['col1', 'col2']].values.tolist()` – Dani Mesejo Jan 03 '19 at 19:55

Convert pandas df columns to paired list

1 Answers1