Converting a dtype: object to datetime not converting

Question

I am trying to convert the df['start_time_time'] column from object to datetime.

start_time	start_time_time
2019-01-01 00:04:37	00:04:37
2019-01-01 00:14:52	00:14:52

The df['start_time_time'] column was extracted from the df['start_time'] column.

I tried converting using this:

pd.to_datetime(df['start_time_time'], format='%H:%M:%S').dt.time

But the data type is still shown as an object.

I also tried this:

pd.to_datetime(df['start_time_time'], format='%H:%M:%S')

Which successfully converts it to datetime64, but the results include the (wrong) date as seen below:

start_time	start_time_time
2019-01-01 00:04:37	1900-01-01 00:04:37
2019-01-01 00:14:52	1900-01-01 00:14:52

Does the datetime format have to include a date?

I am looking for a way to organize/sort the rows by time regardless of date.

Thank you!

Answer 1

Try using this:

df['start_time_time'] = pd.to_datetime(df['start_time_time']).dt.normalize()

It will convert your date to 'datetime64' datatype