Split DF string column and add it as new column

Question

I want to take a DF column and build new column based on str splitting.
Column values looks like that: abcd <> 1234

In order to split i'm using the following:

df['user_id'] = df['Customer User Id'].str.split('<>').str.get(1)

This action works but i'm getting red window saying: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead

Tried to replace the code with loc:

df['user_id'] = df.loc[:,['Customer User Id']]['Customer User Id'].str.split('<>|<').str.get(1)

but the error still shows

Any suggestions..

Is `df` already a slice of another *DataFrame*? Either way, it is only a warning. — pacholik, Dec 16 '15 at 12:40
@RohanAmrute Not sure what data you want to see, the data in this column is strings looks like "abcd <> 1234" and i want to create new column with only the numbers on the right side — mrgold, Dec 16 '15 at 12:46
I tried your code, there is no error or warning of such kind — Rohan Amrute, Dec 16 '15 at 12:46
Possible duplicate of [How to deal with this Pandas warning?](http://stackoverflow.com/questions/20625582/how-to-deal-with-this-pandas-warning) — pacholik, Dec 16 '15 at 12:48
Take look at http://pandas-docs.github.io/pandas-docs-travis/indexing.html#returning-a-view-versus-a-copy and http://stackoverflow.com/q/20625582/1028589 . — pacholik, Dec 16 '15 at 12:50
Added screenshot with the message [link](https://goo.gl/XoSJNl) — mrgold, Dec 16 '15 at 12:52

score 0 · Answer 1 · answered Dec 16 '15 at 13:23

I am using this kind of data:

key1  key2  
22    abcd <> 1234  
34    abcd <> 1234  
12    abcd <> 1234  
55    abcd <> 1234

And the code is:

my_df["key3"] = my_df['key2'].str.split('<>').str.get(1)

Output is :

key1    key2    key3
22  abcd <> 1234    1234
34  abcd <> 1234    1234
12  abcd <> 1234    1234
55  abcd <> 1234    1234

Split DF string column and add it as new column

1 Answers1