pandas column datatype : object to int - Error: float' object has no attribute 'replace'

Question

Sample data in pandas dataframe:

df['Revenue']


0                $0.00
1       $12,982,681.00
2                  NaN
3       $10,150,623.00
4                  NaN
             ...      
1713               NaN
1714               NaN
1715               NaN
1716               NaN
1717               NaN
Name: Revenue, Length: 1718, dtype: object

I need to change the column from currency format to integer so that I can run computations and aggregations.

# Fix format currency
if df['Revenue'].dtype == 'object':

    df['Revenue'] = df['Revenue'].apply(lambda x: x.replace('$','')).apply(lambda x: x.replace(',','')).astype(np.int64)

When I run the above line of code to transform the datatype, I run into the following error:

  3 # Fix format currency
  4 if df['Revenue'].dtype == 'object':
  5     df['Revenue'] = df['Revenue'].apply(lambda x: x.replace('$','')).apply(lambda x: x.replace(',','')).astype(np.int64)
 

  AttributeError: 'float' object has no attribute 'replace'

wwnde · Accepted Answer · 2021-11-12 22:30:15Z

1

You can try replace everything eexcept digits and the dot. If you are reading in file as csv, you can have this controlled at the read stage.

 df['Revenue'].fillna(0).astype(str).replace('[^0-9\.]','', regex=True).str.split('\.').str[0].astype(int)



    Revenue
0            0
1     12982681
2            0
3     10150623
4            0
1713         0
1714         0
1715         0
1716         0
1717         0

edited Nov 12, 2021 at 22:30

answered Nov 12, 2021 at 21:57

wwnde

26.7k6 gold badges22 silver badges38 bronze badges

Sign up to request clarification or add additional context in comments.

9 Comments

kms Over a year ago

how can you control at read stage? can it be done in pd.read_csv?

gshpychka Over a year ago

They don't need the dot either - they're converting to int. You'd also need to do the int conversion at some point.

wwnde Over a year ago

Look at stackoverflow.com/questions/36320317/…

wwnde Over a year ago

@gshpychka he can use the following if he needs them as interger df['Revenue'].astype(str).replace('[^0-9\.]','', regex=True).str.split('\.').str[0]

gshpychka Over a year ago

Use fillna() for those.

|

Collectives™ on Stack Overflow

pandas column datatype : object to int - Error: float' object has no attribute 'replace'

1 Answer 1

9 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

9 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related