add array to pandas data frame

Asked Mar 26 '20 at 14:44

Active Mar 26 '20 at 14:58

Viewed 35 times

I have here an issue and would like to ask for support

Suppose you have a following frame

frame=pd.Dataframe({"Arbitary Number":[1,2,3,4]})

I want to add an additional column, whose entries are np.arrays. I add the entry the following way

frame["new col"]='[8,8,8,8]'

How ever in a later stage I need the entries as array. If I apply

frame["new col"]=frame["new col"].appy(np.array)

I still get object as column type and cannot use the entries to do some math work. I need to go the way with

np.array([eval(xxx)])

to have an array

The question is: Is there a nice and clean way to add arrays as column values without transforming them as strings before assigning them as value? Or if this is not the case and I do need to assign the list as string, is there a way to change the column type to np.array format? My mentioned solution is not working

Thanks a lot for any kind of help

Cheers

asked Mar 26 '20 at 14:44

SMS

that's because your adding an object not an integer based list? try doing `df['new col'].map(eval)` – Umar.H Mar 26 '20 at 14:54
[Convert string representation of list to list](https://stackoverflow.com/questions/1894269/convert-string-representation-of-list-to-list) – Alexandre B. Mar 26 '20 at 14:56
1

`frame["new col"]=[np.array([8,8,8,8])]*frame.shape[0]` – soling Mar 26 '20 at 14:57
exactly what I needed. Thank you very much :) – SMS Mar 26 '20 at 16:22

add array to pandas data frame

0 Answers0