Identify Dates in DataFrame - Pandas-CodePudding

I have a dataframe:

    Datetime
0   2022-06-01 00:00:00 0
1   2022-06-01 00:01:00 0
2   2022-06-01 00:02:00 0
3   2022-06-01 00:03:00 0
4   2022-06-01 00:04:00 0

How to identify the hour is "00", and so for the minutes and seconds. My requirement is to later on like to put them in a function.

CodePudding user response：

You can use:

s = pd.to_datetime(df['Datetime'], format='%Y-%m-%d %H:%M:%S 0') # what is the 0?

df['hour_0'] = s.dt.hour.eq(0)
df['min_0'] = s.dt.minute.eq(0)
df['sec_0'] = s.dt.second.eq(0)

Output:

                Datetime  hour_0  min_0  sec_0
0  2022-06-01 00:00:00 0    True   True   True
1  2022-06-01 00:01:00 0    True  False   True
2  2022-06-01 00:02:00 0    True  False   True
3  2022-06-01 00:03:00 0    True  False   True
4  2022-06-01 00:04:00 0    True  False   True

CodePudding user response：

You can easily extract hours, minutues, seconds directly from date time string. what is extra 0?. If you have extra strings then simply filter first then extra parameters.

df['new'] = pd.to_datetime(df['Datetime'].str.split(' ').str[1],format='%H:%M:%S')
df['hour'] = df['new'].dt.hour
df['minute'] = df['new'].dt.minute
df['second'] = df['new'].dt.second
del df['new']

Gives #

             Datetime      hour  minute  second
0  2022-06-01 00:00:00 0     0       0       0
1  2022-06-01 00:01:00 0     0       1       0
2  2022-06-01 00:02:00 0     0       2       0
3  2022-06-01 00:03:00 0     0       3       0
4  2022-06-01 00:04:00 0     0       4       0

CodePudding user response：

So, your question is a bit unclear to me, but if I understand correctly you just need to extract the hours from your DF? If so the easiest way to do this is to use Pandas inbuilt datetime functionality. For example:

import pandas as pd
df = pd.DataFrame([["2022-12-12 01:59:00"], ["2022-13-12 01:59:00"]])
print(df)

This will yield:

                     0
0  2022-12-12 01:59:00
1   2022-12-13 01:59:00

Now can do:

pd['timestamp'] = pd.to_datetime(df[0])
pd['hour'] = pd['timestamp'].dt.hour

You can do this for minutes and seconds etc. Hope that helps.