CODEWITHSUNDEEP
pythonpandas
Jason Shu
Jason Shu

Reputation: 139

Using pandas.DataFrame.values while getting a different context?

I got a dataframe like

               rectime
0  2018-10-13 00:15:00
1  2018-10-13 00:30:00
2  2018-10-13 00:45:00
3  2018-10-13 01:00:00

The format of 'rectime' is datetime.I'd like to translate it to a list by using

pandas.DataFrame.values

What it returns is something like this

[['2018-10-13T00:15:00.000000000']
 ['2018-10-13T00:30:00.000000000']
 ['2018-10-13T00:45:00.000000000']
 ['2018-10-13T01:00:00.000000000']]

Someone can tell me why there is a difference? And how can I get the list with the same context in 'rectime'?

Upvotes: 0

Views: 38

Answers (1)

andrew_reece
andrew_reece

Reputation: 21274

Use dt.strftime():

df.rectime.dt.strftime("%Y-%m-%d %H:%M:%S").values

array(['2018-10-13 00:15:00', '2018-10-13 00:30:00', '2018-10-13 00:45:00',
       '2018-10-13 01:00:00'], dtype=object)

Explanation: Pandas stores the datetime objects internally in ISO-8601 format, but it renders date strings in a more readable format when you're interacting with printed output.
When you convert out of Pandas into a list with .values, you lose the pretty-print feature, and you get the full 8601 string.
You can use the dt.strftime method to determine exactly how you want the string representation to look, then move it into a list using .values.

Data:

df = pd.DataFrame({'rectime': {
    0: pd.Timestamp('2018-10-13 00:15:00'),
    1: pd.Timestamp('2018-10-13 00:30:00'),
    2: pd.Timestamp('2018-10-13 00:45:00'),
    3: pd.Timestamp('2018-10-13 01:00:00')}})

Upvotes: 2

Related Questions