CODEWITHSUNDEEP
pandasdataframedatetimetimestamp
Juma Hamdan
Juma Hamdan

Reputation: 69

How to convert a pandas column to datetime data type?

I have a dataframe that just has timedate stamps of data type "object". I want to convert the whole dataframe to a datetime data type. Also I would like to convert all the columns to the linux epoch nano seconds. So, I can use this dataframe in pca. enter image description here

Upvotes: 0

Views: 108

Answers (1)

jezrael
jezrael

Reputation: 862761

Sample:

rng = pd.date_range('2017-04-03', periods=3).astype(str)
time_df = pd.DataFrame({'s': rng, 'a': rng})  
print (time_df)
           s          a
0 2017-04-03 2017-04-03
1 2017-04-04 2017-04-04
2 2017-04-05 2017-04-05

Use DataFrame.apply with converting to datetimes and then to native epoch format by convert to numpy array and then to integers:

f = lambda x: pd.to_datetime(x, infer_datetime_format=True).values.astype(np.int64)
#pandas 0.24+
#f = lambda x: pd.to_datetime(x, infer_datetime_format=True).to_numpy().astype(np.int64)
time_df = time_df.apply(f) 
print (time_df)
                     s                    a
0  1491177600000000000  1491177600000000000
1  1491264000000000000  1491264000000000000
2  1491350400000000000  1491350400000000000

Upvotes: 1

Related Questions