如何获取熊猫中两列列之间的日期范围信息答案

【问题标题】：How to get the date range information between two columns columns in pandas如何获取熊猫中两列列之间的日期范围信息
【发布时间】：2020-12-14 13:37:13
【问题描述】：

我有一列日期，分为两列日期，DATE_1 和 DATE_2。我一直在尝试找到一种方法来获取每个日期的周数和该范围内每个日期的星期几，不包括DATE_2 例如：

Date_1      Date_2
2020-09-27  2020-10-01
2020-12-24  2020-12-29
2020-12-24  2021-01-03
2020-12-28  2021-01-03

我想得到

Date_1      Date_2      Week                       Days
2020-09-27  2020-10-01  39,40,40,40                Sun,Mon,Tues,Wed
2020-12-24  2020-12-29  52,52,52,53                Thurs,Fri,Sat,Sun,Mon
2020-12-24  2021-01-03  52,52,52,53,53,53,53,53,53 Thurs,Fri,Sat,Sun,Mon,Tues,Wed,Thur,Fri,Sat
2020-12-28  2021-01-03  53,53,53,53,53,53          Mon,Tues,Wed,Thur,Fri,Sat

天数的显示方式可以是全名，也可以是某天对应的数值，最重要的是数据存在于某处。

我知道 pandas 有 date_range，但我不知道如何将它纳入我正在寻找的内容中。也许它不是特定于熊猫的，我不太确定。任何帮助将不胜感激。

【问题讨论】：

stackoverflow.com/questions/2600775/…的可能重复
这实际上看起来会起作用，您知道是否有一种方法可以扩展列以查找所有未显示日期的 isocalendar() 数据？
看起来这样可以解决问题：stackoverflow.com/questions/7274267/…

标签： python pandas dataframe datetime python-datetime

【解决方案1】：

在 cmets 中的链接的帮助下，我想出了一个使用 date_range 的解决方案：

import pandas as pd

x = {'Date_1': {0: '2020-09-27', 1: '2020-12-24', 2: '2020-12-24', 3: '2020-12-28'},
    'Date_2': {0: '2020-10-01', 1: '2020-12-29', 2: '2021-01-03', 3: '2021-01-03'}}

weekdays = {1: "Mon", 2: "Tues", 3: "Wed", 4: "Thur", 5: "Fri", 6: "Sat", 7: "Sun"}

df = pd.DataFrame(x)

# Creates a new column containing all the days between Date_1 and Date_2
df["Week"] = df.apply(lambda row: pd.date_range(start=row["Date_1"], end=row["Date_2"], freq="D"), axis=1)
# Using the days, we collect the weekdays of the days
df["Days"] = df["Week"].apply(lambda dates: [weekdays.get(date.isocalendar()[2]) for date in dates])
# Finally we gather the week-number for all of the days
df["Week"] = df["Week"].apply(lambda dates: [date.isocalendar()[1] for date in dates])

输出：

       Date_1      Date_2                                          Week                                                        Days
0  2020-09-27  2020-10-01                          [39, 40, 40, 40, 40]                                 [Sun, Mon, Tues, Wed, Thur]
1  2020-12-24  2020-12-29                      [52, 52, 52, 52, 53, 53]                            [Thur, Fri, Sat, Sun, Mon, Tues]
2  2020-12-24  2021-01-03  [52, 52, 52, 52, 53, 53, 53, 53, 53, 53, 53]  [Thur, Fri, Sat, Sun, Mon, Tues, Wed, Thur, Fri, Sat, Sun]
3  2020-12-28  2021-01-03                  [53, 53, 53, 53, 53, 53, 53]                       [Mon, Tues, Wed, Thur, Fri, Sat, Sun]

【讨论】：

【解决方案2】：

你可以这样做：

def f(x):
    didx = pd.date_range(x['Date_1'], x['Date_2'])
    return pd.Series([didx.isocalendar().week.values, 
                      didx.strftime('%a').values], 
                      index=['Weeks', 'Days'])

df[['Weeks', 'Days']] = df.apply(f, axis=1)

输出：

      Date_1     Date_2                                         Weeks                                               Days
0 2020-09-27 2020-10-01                          [39, 40, 40, 40, 40]                          [Sun, Mon, Tue, Wed, Thu]
1 2020-12-24 2020-12-29                      [52, 52, 52, 52, 53, 53]                     [Thu, Fri, Sat, Sun, Mon, Tue]
2 2020-12-24 2021-01-03  [52, 52, 52, 52, 53, 53, 53, 53, 53, 53, 53]  [Thu, Fri, Sat, Sun, Mon, Tue, Wed, Thu, Fri, ...
3 2020-12-28 2021-01-03                  [53, 53, 53, 53, 53, 53, 53]                [Mon, Tue, Wed, Thu, Fri, Sat, Sun]

【讨论】：