Converting Data Type in Pandas

Question

I have a Data Frame with 15 columns suppose out of which i want only 6. I am performing aggregate and then group by but it is throwing error.

def my_compute_function(my_input):

    df=pd.DataFrame(my_input)
    df2=df[(df['D'] == "Validated")]
    df2[['A','E','F']]=df2[['A','E','F']].apply(pd.to_numeric) 
    df3=df2[['A','B','C','D','E','F']].groupby(['B','C','D']).agg({'A': 
   'max','E': 'max','F': 'max'}).reset_index()

return df3

So i want only 6 columns A,B,C,D,E,F.
When i am adding this line

df2[['A','E','F']]=df2[['A','E','F']].apply(pd.to_numeric)

it is throwing error that

ValueError: can not infer schema from empty dataset.

Are you getting an actual indentation error or another error? Please [edit] your post to include the *full error* — Jon Clements, Aug 16 '17 at 07:47
IndentationError: Unexpected Indent. I also checked the spaces its fine, and when i am removing this line my code is working fine. — vatsal, Aug 16 '17 at 07:52
See this post for more guidance https://stackoverflow.com/a/45621723/2336654 — piRSquared, Aug 16 '17 at 07:58
please check the edited question. Earlier i was getting indentation error but when i used df3 then i am getting the above error. — vatsal, Aug 16 '17 at 12:27

score 3 · Answer 1 · answered Aug 16 '17 at 07:50

This should work.

df[['A', 'B', 'C']] = df[['A', 'B', 'C']].apply(pd.to_numeric)

This will break

# Ignore this comment, I needed SO to show a blank
# line followed by code indented by a space

 df[['A', 'B', 'C']] = df[['A', 'B', 'C']].apply(pd.to_numeric)

jezrael · Answer 2 · 2017-08-16T07:55:34.023

1

I think you can copy whitespaces from working code before problematic line:

#some code before, copy whitespaces from starts of line to first letter
    df = pd.read_csv(file)
^^^^copy here

    df[['A','B','C']]=df[['A','B','C']].apply(pd.to_numeric)  
^^^^paste here

edited Aug 16 '17 at 07:55

answered Aug 16 '17 at 07:46

jezrael

822,522
95
1,334
1,252

still same error IndentationError: Unexpected Indent. – vatsal Aug 16 '17 at 07:50
when i am assigning this Data Frame to another Data Frame then my code is executing fine but then it is giving error - cannot infer to an empty dataset. – vatsal Aug 16 '17 at 08:25
Is possible see your code? And if possible, the best is send me your file to my email (in my profile) for check this problematic line. Without it is hard find problem connected with indent error. thanks. – jezrael Aug 16 '17 at 08:27

Converting Data Type in Pandas

2 Answers2