Python pandas: convert a column to a string?

Question

I have some columns:

0 ['XS', '2X'] 1 ['A2', '2X', '8W'] 2 ['2X'] 3 ['A2', 'FR', 'XS', '8W', '8E', '2X'] 4 ['A5', 'FR', 'XS', 'QF', '2X', '23'] 5 ['XS', '2X', 'MF'] 6 None

that I am trying to iterate over. It I loop over the column and get the type:

for item in data1['business_types']: print(type(item))

It shows that the types are of type str

<class 'str'> <class 'str'> <class 'str'> <class 'str'> <class 'str'> <class 'str'> <class 'NoneType'>

I would like to convert the objects in this column to a list object that I can iterate over.

Maybe l = data1['business_types'].tolist() and then l = [ast.literal_eval(i) for i in l] — user2918461
– user2918461, Commented Jun 7, 2017 at 13:04
I think your title should be convert column of strings to column of lists of strings — user2918461
– user2918461, Commented Jun 7, 2017 at 13:05

jezrael · Accepted Answer · 2017-06-07 13:34:43Z

I think you need ast.literal_eval with fillna for replace None to []:

data1['business_types'] = data1['business_types'].fillna('[]').apply(ast.literal_eval) print (data1) business_types 0 [XS, 2X] 1 [A2, 2X, 8W] 2 [2X] 3 [A2, FR, XS, 8W, 8E, 2X] 4 [A5, FR, XS, QF, 2X, 23] 5 [XS, 2X, MF] 6 []

If need None (NaNs):

def f(x): try: return ast.literal_eval(x) except ValueError: return x import ast data1['business_types'] = data1['business_types'].apply(f) print (data1) business_types 0 [XS, 2X] 1 [A2, 2X, 8W] 2 [2X] 3 [A2, FR, XS, 8W, 8E, 2X] 4 [A5, FR, XS, QF, 2X, 23] 5 [XS, 2X, MF] 6 NaN for item in data1['business_types']: print(type(item)) <class 'list'> <class 'list'> <class 'list'> <class 'list'> <class 'list'> <class 'list'> <class 'float'>

What version are you using? I actually had tried that and I get ValueError: malformed node or string: None under 0.20.2. I actually just updated from 0.19.2 and it's broke some of my code...
I use pandas 0.20.1, but if bad data, then it can failed :(
This worked, I had tried a solution with ast.literal_eval function however, the trick with filling the Nonetypes with [] was clever and worked. Thanks.

Collectives™ on Stack Overflow

Python pandas: convert a column to a string?

1 Answer 1

6 Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

6 Comments

Related