fra_P
fra_P

Reputation: 23

Add date to an DataArray of float (or viceversa)

I'm kind of new in python. I have an xarray DataArray which contains a variable made of floats. These floats are supposed to be seconds. I would like to add these floats to a given date:

2021-01-01 00:00:00 + 1 =  2021-01-01 00:00:01
...
2021-01-01 00:00:00 + 120 = 2021-01-01 00:02:00

in order have an array of datetime I can work with (for example to extract the events between a given interval of time)

I tried to use timedelta but it does not like DataArray. Is there a workaround for that?

Upvotes: 2

Views: 766

Answers (2)

spencerkclark
spencerkclark

Reputation: 2097

For standard calendar operations, xarray works best with NumPy datetime64 and timedelta64 types. In your particular case, I would recommend converting your DataArray of floats to timedelta64 (taking into account the units of seconds), and adding those values to a datetime64 instance. This will produce a DataArray of type datetime64[ns], which you can use in other operations:

In [1]: import numpy as np; import xarray as xr

In [2]: data = np.arange(10.)

In [3]: da = xr.DataArray(data, dims=["x"])

In [4]: np.datetime64("2000-01-01") + da.astype("timedelta64[s]")
Out[4]:
<xarray.DataArray (x: 10)>
array(['2000-01-01T00:00:00.000000000', '2000-01-01T00:00:01.000000000',
       '2000-01-01T00:00:02.000000000', '2000-01-01T00:00:03.000000000',
       '2000-01-01T00:00:04.000000000', '2000-01-01T00:00:05.000000000',
       '2000-01-01T00:00:06.000000000', '2000-01-01T00:00:07.000000000',
       '2000-01-01T00:00:08.000000000', '2000-01-01T00:00:09.000000000'],
      dtype='datetime64[ns]')
Dimensions without coordinates: x

Upvotes: 2

Jake Cubes
Jake Cubes

Reputation: 7

This is for time:

from datetime import datetime

now = datetime.now()

current_time = now.strftime("%H:%M:%S")

print(current_time)

This is for date:

from datetime import date

today = date.today()

print("Today's date:", today)

Upvotes: -1

Related Questions