How to make a new column of numpy arrays in a pandas data frame?

Question

I have the following data frame:

df = pd.DataFrame({'id': [1, 2, 3, 4],
                   'a': ['on', 'on', 'off', 'off'],
                   'b': ['on', 'off', 'on', 'off']})

How can I create a new column df['new'] with the type of NumPy arrays/lists so that I can perform operations like:

df.loc[1, 'new'] = np.array([2 , 'l'])
#or
df.loc[1, 'new'] = [2 , 'l']

Do you really want arrays as opposed to lists? An object dtype array column can hold anything - array, list, dict, string, None` — hpaulj
– hpaulj, Commented Aug 18, 2021 at 19:20
Why do you need to "declare" the type of object you are putting in a dataframe? Just put it there. — Aryerez
– Aryerez, Commented Aug 18, 2021 at 19:21
@Aryerez I thought so at first, but I was getting the following error when I do it directly:ValueError: cannot set using a multi-index selection indexer with a different length than the value — Kaihua Hou
– Kaihua Hou, Commented Aug 18, 2021 at 19:23

not_speshal · Accepted Answer · 2021-08-18 19:28:49Z

2

Specify the datatype as "object" while creating the new column and then insert the elements as needed:

df["new"] = pd.Series(dtype="object")
df.at[1, 'new'] = [2 , 'l']
>>> df
   id    a    b     new
0   1   on   on     NaN
1   2   on  off  [2, l]
2   3  off   on     NaN
3   4  off  off     NaN

answered Aug 18, 2021 at 19:28

not_speshal

23.2k2 gold badges18 silver badges33 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

How to make a new column of numpy arrays in a pandas data frame?

1 Answer 1

Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Related