Pandas Lambda Function : attribute error 'occurred at index 0'

Question

I am using Pandas to create a new column in a data frame created from a csv.

[in] DfT_raw = pd.read_csv('./file.csv', index_col = False)
[in] print(DfT_raw)

[out]            Region Name dCount ONS    CP  S Ref E  S Ref N   Road  \
0        East Midlands  E06000015      14/04/00 00:00  37288   434400   336000   A516   
1        East Midlands  E06000015       14/04/00 00:00  37288   434400   336000   A516   
2        East Midlands  E06000015       14/04/00 00:00  37288   434400   336000   A516   
3        East Midlands  E06000015       14/04/00 00:00  37288   434400   336000   A516

I define a function to strip the time from the datetime fieldn (dCount) and then create a new column 'date'

[in] def date_convert(dCount):
         return dCount.date()

     DfT_raw['date'] = DfT_raw.apply(lambda row: date_convert(row['dCount']), axis=1)

[out] AttributeError: ("'str' object has no attribute 'date'", u'occurred at index 0')

There is some issue with the index_col. I previously used index_col = 1 but got the same error.

When I print 'dCount' I get

0          14/04/00 00:00
1          14/04/00 00:00
2          14/04/00 00:00
3          14/04/00 00:00
4          14/04/00 00:00

The index column is causing the error. How do I ensure this isn't given to the function?

It's a str not a datetime object, you need to convert first df['dCount'] = pd.to_datetime(df['dCount']) and then get the date df['dCount'].dt.date, you could have read those datetime strings in as datetime by passing parse_dates=['dCount'] to read_csv — EdChum
– EdChum, Commented Oct 14, 2015 at 9:02
DfT_raw = pd.read_csv('./file.csv', parse_dates=['dCount'],index_col = False) works perfect, many thanks! Feel free to post as answer if you wish — LearningSlowly
– LearningSlowly, Commented Oct 14, 2015 at 10:43

EdChum · Accepted Answer · 2015-10-14 10:44:42Z

4

Your error here is that your dates are str not datetime, either convert using to_datetime:

df['dCount'] = pd.to_datetime(df['dCount'])

or better just tell read_csv to parse that column as datetime:

DfT_raw = pd.read_csv('./file.csv', parse_dates=['dCount'],index_col = False)

Afterwards you can then get just the date by calling the dt.date accessor

answered Oct 14, 2015 at 10:44

EdChum

397k204 gold badges836 silver badges583 bronze badges

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

Collectives™ on Stack Overflow

Pandas Lambda Function : attribute error 'occurred at index 0'

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related