combine one string column to another with Pandas

Question

If i have a following dataframe:

id     categories      products
01        fruit       Apple, Apricot
02        fruit       Apple, Banana, Clementine, Pear
03        fruit       Orange, Pineapple, Pear
04      vegetable     Carrot, Cabbage

and i want create a new df like this, what i should do? thanks.

id          products
01     (fruit)Apple, Apricot
02     (fruit)Apple, Banana, Clementine, Pear
03     (fruit)Orange, Pineapple, Pear
04     (vegetable)Carrot, Cabbage

score 4 · Answer 1 · answered Jun 27 '18 at 01:16

4

Just sum strings

"(" + df.categories + ")" + df.products

answered Jun 27 '18 at 01:16

rafaelc

57,686
15
58
82

do you know why `df['format'] = '({0}){1}'.format(df.categories, df.products)` doesn't work in that way? – aydow Jun 27 '18 at 02:08
1

@aydow because you will be getting the result of the conversion of the arguments (in this case, the `Series` objects `df.categories` and `df.products`) to `string` and replacing it in the` {0}` and `{1}`. The result will be a single string and not a series as you probably intended – rafaelc Jun 27 '18 at 02:23

Dickster · Accepted Answer · 2018-06-27T01:19:26.947

0

try this

    df = df.set_index('id')
    df  =df.apply(lambda x: "(" + x['categories'] + ")" + x['products'],axis=1).to_frame('products')
    print df

edited Jun 27 '18 at 01:19

answered Jun 27 '18 at 01:13

Dickster

2,969
3
23
29

At this point, why not just `'(' + x.categories + ')' + x.products`? – hilberts_drinking_problem Jun 27 '18 at 01:15
to maintain 'id' – Dickster Jun 27 '18 at 01:17
You can still do that; it seems that the direct approach is both simpler and more efficient than using `apply`. – hilberts_drinking_problem Jun 27 '18 at 01:19

combine one string column to another with Pandas

2 Answers2