Python Forum

I have a Pandas Dataframe where two columns contain times (stored as strings).

I want to get the minimum difference (hours, minutes, seconds) between both columns (Column1 and Column2).

I have converted both columns to times as per below and stored them in different dataframes:

Column1Times = pd.to_datetime(myDataFrame['Column1'], format='%H:%M:%S').dt.time
Column2Times = pd.to_datetime(myDataFrame['Column2'], format='%H:%M:%S').dt.time

To get the minimum time difference I use below, but I get

Error:
operand "-" not suuported.

print("Minimum difference is : ", (Column1Times - Column2Times).min())

Can someone please tell me what I am doing wrong?

You can only find differences for datetime objects, try this:

(Column1Times - Column2Times).apply(lambda x: abs(x.days * 86400 * 10**6 + x.seconds * 10**6 + x.microseconds)%(86400 * 10**6) / 10**6) 
# note absolute values used (result in seconds)

I tried following:

>>> import pandas as pd
>>> Column1Times = pd.to_datetime('10:10:10', format='%H:%M:%S')
>>> Column2Times = pd.to_datetime('11:12:13', format='%H:%M:%S')
>>> delta = Column2Times - Column1Times
>>> delta
Timedelta('0 days 01:02:03')
>>> delta.seconds
3723
>>> delta.total_seconds()      
3723.0
>>> delta.components
Components(days=0, hours=1, minutes=2, seconds=3, milliseconds=0, microseconds=0, nanoseconds=0)
>>> delta.components.hours, delta.components.minutes, delta.components.seconds
(1, 2, 3)

I am trying to remember why I did create so sophisticated answer, but no success; In any case, I think that the expected by the OP answer would be something like min(delta.seconds, (-delta).seconds), where delta is difference between two datetime objects. So, the difference between two times 23:59:00 and 00:00:01 should be 61 sec.

(Mar-21-2019, 10:54 AM)scidam Wrote: [ -> ]I am trying to remember why I did create so sophisticated answer, but no success;

Been there, done that