Converting time-stamps in the dataframe

Question

Thanks to this answer, I have the following dataframe:

              START_POINT   END_POINT   DURATION

island Stage            
1   SLEEP-S0    00:32:03    00:42:33    630.0
2   SLEEP-S1    00:42:33    00:45:03    150.0
3   SLEEP-S0    00:45:03    00:46:03    60.0
4   SLEEP-S1    00:46:03    00:48:33    150.0
5   SLEEP-S2    00:48:33    00:50:03    90.0
... ... ... ... ...
127 SLEEP-S2    09:32:03    09:39:03    420.0
128 SLEEP-S0    09:39:03    09:39:33    30.0
129 SLEEP-S1    09:39:33    09:40:03    30.0
130 SLEEP-S2    09:40:03    09:48:03    480.0
131 SLEEP-S0    09:48:03    NaN NaN

However, I want to convert the times here into times (in float or int) starting from t = 0. For example, this is what I want:

               START_POINT   END_POINT  DURATION

island Stage            
1   SLEEP-S0    0                630    630.0
2   SLEEP-S1    630              780    150.0
3   SLEEP-S0    780              840    60.0
4   SLEEP-S1    ...             ...     ...
5   SLEEP-S2    ...             ...    ...
... ... ... ... ...

and so on. Can someone please help?

Asked By: S C

||

Source

Answer 1

Reference this answer below
How to standardize/normalize a date with pandas/numpy?

You could standardize the array of timestamps and then you could multiply the standardized values by 10000000 to convert to an int.

Answered By: jpetrantoni

Answer 2

here you go:

df["start"] = df["DURATION"].cumsum().shift().fillna(0)
df["end"] = df["DURATION"].cumsum()

df
Out[5]: 
                START_POINT END_POINT  DURATION   start     end
island Stage                                                   
1      SLEEP-S0    00:32:03  00:42:33     630.0     0.0   630.0
2      SLEEP-S1    00:42:33  00:45:03     150.0   630.0   780.0
3      SLEEP-S0    00:45:03  00:46:03      60.0   780.0   840.0
4      SLEEP-S1    00:46:03  00:48:33     150.0   840.0   990.0
5      SLEEP-S2    00:48:33  00:50:03      90.0   990.0  1080.0
127    SLEEP-S2    09:32:03  09:39:03     420.0  1080.0  1500.0
128    SLEEP-S0    09:39:03  09:39:33      30.0  1500.0  1530.0
129    SLEEP-S1    09:39:33  09:40:03      30.0  1530.0  1560.0
130    SLEEP-S2    09:40:03  09:48:03     480.0  1560.0  2040.0
131    SLEEP-S0    09:48:03       NaN       NaN  2040.0     NaN

Answered By: FObersteiner

Converting time-stamps in the dataframe

Question:

Answers: