Datetime of pandas time series preamble

「 This is my participation 11 The fourth of the yuegengwen challenge 26 God , Check out the activity details ：2021 One last more challenge 」

Before we study time series, we need to understand datetime Basic use of modules ,datetime Modules are not pandas The library contains .

But in order to learn better pandas Time series in , We can learn first datetime In order to lay a good foundation ,datetime We mainly master several methods ：datetime.date(), datetime.datetime(), datetime.timedelta()

datetime.date

We can use datetime.date.today Method returns the current time , Its data type is datetime.date

import datetime  
today = datetime.date.today()
print(today,type(today))
>>>
2018-10-28 <class 'datetime.date'>  Copy code 

At the same time, we can use datetime.date( year , month , Japan ) Output the date we want datetime type .

import datetime
t = datetime.date(2016,6,1)
print(t)
# ( year , month , Japan ) →  Get the date directly 
>>>
2016-06-01
 Copy code 

datetime.datetime

We use datetime.datetime.now Method returns the current time , Its data type is datetime.datetime

import datetime
now = datetime.datetime.now()
print(now,type(now))
>>>
2018-10-28 17:34:07.467642 <class 'datetime.datetime'>  Copy code 

In the same way, we can use datetime.datetime( year , month , Japan , when , branch , second ) Output what we want in the form of datetime.datetime Data of type .

t1 = datetime.datetime(2016,6,1)
t2 = datetime.datetime(2014,1,1,12,44,33)
print(t1,t2)
>>>
2016-06-01 00:00:00 2014-01-01 12:44:33
 Copy code 

datetime.timedelta

datetime.timedelta The data represented is the time difference data type . Time difference is mainly used as the addition and subtraction of time , Equivalent to recognizable time “ Difference value ”

# datetime.timedelta： Time difference 

today = datetime.datetime.today()  # datetime.datetime Also have today() Method 
yestoday = today - datetime.timedelta(1)  #
print(today)
print(yestoday)
print(today - datetime.timedelta(7))
>>>
2018-11-04 14:48:28.967149
2018-11-03 14:48:28.967149
2018-10-28 14:48:28.967149
 Copy code 

Date resolution method ：parser.parse

In addition to the above conversion method , We also need other methods for parsing strings into time format .

have access to parse Method to convert a string to datetime.datetime Data of type .

from dateutil.parser import parse
date = '12-21-2017'
t = parse(date)
print(t,type(t))
>>>
2017-12-21 00:00:00 <class 'datetime.datetime'>  Copy code 

Similarly, we can also convert strings in other formats into datetime.datetime, But I can't parse Chinese

print(parse('2000-1-1'),'\n',
     parse('5/1/2014'),'\n',
     parse('5/1/2014', dayfirst = True),'\n',  
     #  In the international common format , The day is before the month , Can pass dayfirst To set up 
     parse('22/1/2014'),'\n',
     parse('Jan 31, 1997 10:45 PM'))
>>>
2000-01-01 00:00:00
 2014-05-01 00:00:00
 2014-01-05 00:00:00
 2014-01-22 00:00:00
 1997-01-31 22:45:00
 Copy code 

Pandas Time data ：Timestamp

• The time data represents the point in time , yes pandas Data type of
• It is the most basic type of time series data that associates values with time points .

Timestamp And the above datetime Basically no difference , The only difference from one is pandas modular , One is datetime modular .

pandas.Timestape

pandas.Timestape Can generate directly pandas Time data

import numpy as np
import pandas as pd
date1 = datetime.datetime(2016,12,1,12,45,30)  #  Create a datetime.datetime
date2 = '2017-12-21'  #  Create a string 
t1 = pd.Timestamp(date1)
t2 = pd.Timestamp(date2)
print(t1,type(t1))
print(t2,type(t2))
>>>
2016-12-01 12:45:30 <class 'pandas._libs.tslibs.timestamps.Timestamp'> 2017-12-21 00:00:00 <class 'pandas._libs.tslibs.timestamps.Timestamp'>  Copy code 

pandas.to_datetime

pandas.to_datetime If it is a single time data , convert to pandas Time data , The data type is Timestamp, If it is multiple time data , Will be converted to pandas Of DatetimeIndex.

Single time data instance ：

from datetime import datetime

date1 = datetime(2016,12,1,12,45,30)
date2 = '2017-12-21'
t1 = pd.to_datetime(date1)
t2 = pd.to_datetime(date2)
print(t1,type(t1))
print(t2,type(t2))
>>>
2016-12-01 12:45:30 <class 'pandas._libs.tslibs.timestamps.Timestamp'> 2017-12-21 00:00:00 <class 'pandas._libs.tslibs.timestamps.Timestamp'>  Copy code 

Multiple time data instances ：

lst_date = [ '2017-12-21', '2017-12-22', '2017-12-23']
t3 = pd.to_datetime(lst_date)
print(t3,type(t3))
>>>
DatetimeIndex(['2017-12-21', '2017-12-22', '2017-12-23'], dtype='datetime64[ns]', freq=None) <class 'pandas.core.indexes.datetimes.DatetimeIndex'>  Copy code 

When there are other data formats in multiple time series , We can use error Parameter return ,errors = 'ignore': Returns the original input when it is not parsed , Here is the direct generation of general arrays

date3 = ['2017-2-1','2017-2-2','2017-2-3','hello world!','2017-2-5','2017-2-6']
t3 = pd.to_datetime(date3, errors = 'ignore')
print(t3,type(t3))
>>>
['2017-2-1' '2017-2-2' '2017-2-3' 'hello world!' '2017-2-5' '2017-2-6'] <class 'numpy.ndarray'>  Copy code 

When errors = 'coerce' when , Data of different data types will be returned NaT, The result is that DatetimeIndex

date3 = ['2017-2-1','2017-2-2','2017-2-3','hello world!','2017-2-5','2017-2-6']
t4 = pd.to_datetime(date3, errors = 'coerce')
print(t4,type(t4))
>>>
DatetimeIndex(['2017-02-01', '2017-02-02', '2017-02-03', 'NaT', '2017-02-05',
               '2017-02-06'],
              dtype='datetime64[ns]', freq=None) <class 'pandas.core.indexes.datetimes.DatetimeIndex'>  Copy code