Pandas – Dataframe dates subtraction

Question

I am dealing with a dataframe like this:

mydata['TS_START']

0        2022-11-09 00:00:00
1        2022-11-09 00:00:30
2        2022-11-09 00:01:00
3        2022-11-09 00:01:30
4        2022-11-09 00:02:00
...

I would like to create a new column where:

mydata['delta_t']
0        2022-11-09 00:00:30 - 2022-11-09 00:00:00
1        2022-11-09 00:01:00 - 2022-11-09 00:00:30
2        2022-11-09 00:01:30 - 2022-11-09 00:01:00
3        2022-11-09 00:02:00 - 2022-11-09 00:01:30
...

Obtaining something like this (in decimals units hour based):

mydata['delta_t']
0        30/3600
1        30/3600
2        30/3600
3        30/3600
...

I obtained this result using a for cycle, but it is very slow. I would like to obtain a faster solution, using a vectorized form.
Do you have any suggestion?

Asked By: niccolo_guazz

||

Source

Answer 1

here is one way :

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

df['delta_t'] = (df['date'] - df['date'].shift(1)).dt.total_seconds()
print(df)

output :

>>
                 date  delta_t
0 2022-11-09 00:00:00      NaN
1 2022-11-09 00:00:30     30.0
2 2022-11-09 00:01:00     30.0
3 2022-11-09 00:01:30     30.0
4 2022-11-09 00:02:00     30.0

Answered By: eshirvana

Pandas – Dataframe dates subtraction

Question:

Answers: