Python dataframe list-column splitting

Question

I have a dataframe with one of the columns being a list like so:

RefID	Ref
Ref1	(baby, 60, 0)
Ref2	(something, 90, 2)

I wanted to extract this list as separate fields, as in this code:

df['MatchWord'], df['Prox'], df['MatchID'] = df.Ref.str

But I get the "FutureWarning: Columnar iteration..." error. How else can I split this list-column?

Is this something that you are looking for? https://stackoverflow.com/questions/35491274/split-a-pandas-column-of-lists-into-multiple-columns — Mayur Kr. Garg, Jan 23 '22 at 21:21
@Mayur Kr. Garg , no that is for arrays (I forget python terminology) and this is already a list, to I can't use _tolist — user17983408, Jan 23 '22 at 21:34

mozway · Answer 1 · 2022-01-23T21:40:59.253

1

You can apply(pd.Series). This will unpack the items as columns:

df['Ref'].apply(pd.Series, index=['MatchWord', 'Prox', 'MatchID'])

Or use the DataFrame constructor:

pd.DataFrame(df['Ref'].to_list(), columns=['MatchWord', 'Prox', 'MatchID'])

Output:

   MatchWord  Prox  MatchID
0       baby    60        0
1  something    90        2

edited Jan 23 '22 at 21:40

answered Jan 23 '22 at 21:35

mozway

194,879
13
39
75

Thanks, this is great. works perfectly – user17983408 Jan 23 '22 at 22:36

score 1 · Accepted Answer · answered Jan 23 '22 at 21:42

1

Use pd.DataFrame.from_records and join:

cols = ['MatchWord', 'Prox', 'MatchID']
out = df.join(pd.DataFrame.from_records(df['Ref'], index=df.index, columns=cols))
print(out)

# Output
  RefID                 Ref  MatchWord  Prox  MatchID
0  Ref1       (baby, 60, 0)       baby    60        0
1  Ref2  (something, 90, 2)  something    90        2

answered Jan 23 '22 at 21:42

Corralien

109,409
8
28
52

Nice alternative ;) – mozway Jan 23 '22 at 22:06
Thanks. This works well, and the both of you have given me some alternatives to play with! – user17983408 Jan 23 '22 at 22:35

Python dataframe list-column splitting

2 Answers2