CODEWITHSUNDEEP
pythonpandasdatetimetype-conversiondtype
snowballtrader
snowballtrader

Reputation: 43

Pandas converting and combining object, integer columns to datetime

Suppose I have following data with data formats:

date time
2022-04-21 610
2022-04-22 610
2022-04-22 1751

date: object

time: int32

I wish to combine date column and time column to make new column 'datetime' with datetime format like this:

date time datetime
2022-04-21 610 2022-04-21 6:10
2022-04-22 610 2022-04-22 7:01
2022-04-22 1751 2022-04-22 17:51

When I use following codes, I get errors as following:

  1. Converting object to datetime

    df['date'] = pd.to_datetime(df['date'], format='%Y-%M-%d')

date time
2022-01-21 00:04:00 610
2022-01-22 00:04:00 610
2022-01-22 00:04:00 1751

  1. Converting int32 to datetime

    df_trade['time'] = pd.to_datetime(df_trade['time'], format='%H:%M')

ValueError: time data '610' does not match format '%H:%M' (match)

I guess the last step must be easy with following code, but columns are not in the right format to combine.

df_trade['datetime'] = df_trade['date'].astype(str)+" "+df_trade['time'].astype(str)

How do I convert the object, int32 columns into datetime format without changing the contents?

Upvotes: 0

Views: 403

Answers (1)

jezrael
jezrael

Reputation: 863266

You are close, all your solution should be combined, only change few typos - for months need %m and for parse times need remove : - %H%M:

df_trade['datetime'] = pd.to_datetime(df_trade['date'] +' '+ df_trade['time'].astype(str), 
                                      format='%Y-%m-%d %H%M')
print (df_trade)
         date  time            datetime
0  2022-04-21   610 2022-04-21 06:10:00
1  2022-04-22   610 2022-04-22 06:10:00
2  2022-04-22  1751 2022-04-22 17:51:00

Upvotes: 2

Related Questions