在python中，如何在带有dtype对象的列上使用split('_')？

Question

I am working on a data frame whose one column is of type object:

example: name 36512 non-null object

I have tried the following with no success

> name_str = autos['name'].to_string()
> print(type(name_str))   # this makes the name_str as type string
> autos['name'] = name_str  # putting it back in the data frame brought it back to type object

Also, tried the following:

> import json
> autos['name'] = json.dumps(name_str)

My goal is to split the first two words using split('_') , but unable to do so unless the type is string

example: BMW_740i_4_4_Liter_HAMANN_UMBAU_Mega_Optik

output: ['BMW', '740i'] in a new column

Answer 1

import pandas as pd

df = pd.DataFrame({'name':['BMW_740i_4_4_Liter_HAMANN_UMBAU_Mega_Optik']})

df['new'] = df['name'].str.split('_').str[:2]

print(df)

Output:

                                         name          new
0  BMW_740i_4_4_Liter_HAMANN_UMBAU_Mega_Optik  [BMW, 740i]

Note that a column of strings is considered an "object" dtype in pandas so you should have the right dtype already