【数据分析&数据挖掘】pandas时间数据

 1 import pandas as pd
 2 """
 3 pandas默认支持的时间点类型——Timestamp
 4 pandas默认支持的时间序列类型——DatetimeIndex
 5 numpy默认支持的时间点数据类型——datetime64
 6 """
 7
 8 # 可以使用pd.to_datetime 将时间点转化为pandas默认支持的时间点类型
 9 res = pd.to_datetime("2019-11-11")
10 print("res: \n", res)
11 print("res的类型: \n", type(res))
12
13 # 可以使用pd.to_datetime 将时间序列转化为pandas支持的时间序列类型
14 res = pd.to_datetime(["2019-11-11", "2019-12-12", "2020-02-14", "2020-03-07"])
15 print("res: \n", res)
16 print("res的类型: \n", type(res))
17
18 # 可以使用pd.DatetimeIndex 将时间序列转化为pandas支持的时间序列类型, 不能转化时间点
19 res = pd.DatetimeIndex(["2019-11-11", "2019-12-12", "2020-02-14", "2020-03-07"])
20 print("res: \n", res)
21 print("res的类型: \n", type(res))
22
23 # 加载detail
24 detail = pd.read_excel("../day05/meal_order_detail.xlsx")
25 print("detail: \n", detail)
26 print("detail的列名称: \n", detail.columns)
27 print(detail.dtypes)
28
29 # 将 place_order_time 转化为pandas默认支持的时间序列类型
30 detail.loc[:, "place_order_time"] = pd.to_datetime(detail.loc[:, "place_order_time"])
31 print(detail.dtypes)
32 # 可以提取出时间序列中的属性
33
34 # 年属性
35 year = [i.year for i in detail.loc[:, "place_order_time"]]
36 print("year: \n", year)
37
38 # 月属性
39 month = [i.month for i in detail.loc[:, "place_order_time"]]
40 print("month: \n", month)
41
42 # 日属性
43 day = [i.day for i in detail.loc[:, "place_order_time"]]
44 print("day: \n", day)
45
46 # 周属性——一年的第N周
47 week = [i.week for i in detail.loc[:, "place_order_time"]]
48 print("week: \n", week)
49
50 week_of_year = [i.weekofyear for i in detail.loc[:, "place_order_time"]]
51 print("week_of_year: \n", week_of_year)
52
53 day_of_year = [i.dayofyear for i in detail.loc[:, "place_order_time"]]
54 print("day_of_year: \n", day_of_year)
55
56 # 获取一周中的第N天
57 day_of_week = [i.dayofweek for i in detail.loc[:, "place_order_time"]]
58 print("day_of_week: \n", day_of_week)
59
60 # 获取周几
61 weekday = [i.weekday for i in detail.loc[:, "place_order_time"]]
62 print("weekday: \n", weekday)
63
64 weekday_name = [i.weekday_name for i in detail.loc[:, "place_order_time"]]
65 print("weekday_name: \n", weekday_name)
66
67 # 获取第几季度
68 quarter = [i.quarter for i in detail.loc[:, "place_order_time"]]
69 print("quarter: \n", quarter)
70
71 # 时间数据的运算
72 res = pd.to_datetime("2019-11-11") + pd.Timedelta(days=2)
73 res = pd.to_datetime("2019-11-11") + pd.Timedelta(weeks=1)
74 res = pd.to_datetime("2019-11-11") + pd.Timedelta(weeks=-1)
75
76 # 时间差——返回days
77 res = pd.to_datetime("2019-11-11") - pd.to_datetime("2002-1-8")
78 print("res: \n", res)
79 res = res.days
80 print("res: \n", res)
81 res = res/365
82 print("年龄: \n", res)
83
84 # 还可以获取本机的最初始时间、最大时间
85 print("本机的最小时间: \n", pd.Timestamp.min)
86 print("本机的最大时间: \n", pd.Timestamp.max)
87
88 # 生成时间数据的API
89 # start——开始日期
90 # end——结束日期
91 # periods——如果end不传， 生成时间数据的数量
92 # freq——默认按天
93 res = pd.date_range(start="2019-11-11", periods=5)
94 res = pd.date_range(start="2019-11-11", end="2019-11-16")  # end和period不能同时传
95 # 生成频次为36天
96 res = pd.date_range(start="2019-11-11", end="2020-11-16", freq="36D")
97 print(res)

原文地址：https://www.cnblogs.com/Tree0108/p/12116054.html