CODEWITHSUNDEEP
pythonpandasdatetime
pepe
pepe

Reputation: 9909

How to create new Pandas column with total daily minutes calculated from timestamp?

Given the following df in which timestamps are not index:

   timestamp
0  2020-10-23 12:20:00-04:00  
1  2020-10-23 12:30:00-04:00  
2  2020-10-23 12:40:00-04:00
3  2020-10-23 12:50:00-04:00

I am trying to create a new column minutes that contains the total number of minutes as counted from 00:00:00.

So the above would output

   timestamp                      minutes
0  2020-10-23 12:20:00-04:00      740
1  2020-10-23 12:30:00-04:00      750
2  2020-10-23 12:40:00-04:00      760
3  2020-10-23 12:50:00-04:00      770

I've been trying pd.timedelta however I'm unable to set a begin time to reference from (in my case I'd want the minutes counted from 00:00:00 hs). Importantly, minutes should be integer and not string.

Any suggestions on how to tackle this?

Upvotes: 1

Views: 37

Answers (1)

Quang Hoang
Quang Hoang

Reputation: 150735

You can use dt.normalize to get the day, subtract, and divide the timdelta 1T:

df['minutes'] = (df.timestamp - df.timestamp.dt.normalize()) // pd.Timedelta('1T')

Another option (obviously easy):

df['minutes'] = df.timestamp.dt.hour * 60 + df.timestamp.dt.minute

Output:

                  timestamp  minutes
0 2020-10-23 12:20:00-04:00      740
1 2020-10-23 12:30:00-04:00      750
2 2020-10-23 12:40:00-04:00      760
3 2020-10-23 12:50:00-04:00      770

Upvotes: 1

Related Questions