Python：根据 pandas 数据框中的每日数据计算周开始和周结束？

Question

我有不同月份的每日数据集。我想根据每种产品类型和国家/地区计算周开始（星期日）和周末（星期六），值应该是该特定周的平均值。

dates                   product        country      value  name

2021-10-01 00:00:00     Voice           Lucia       2      A
2021-10-01 00:00:00     TV              Jamai       1      A
2021-10-01 00:00:00     TV              Trin        5      A
2021-10-01 00:00:00     Voice           Gren        5      A
2021-10-01 00:00:00     Broad           Vin         7      A
2021-10-01 00:00:00     TV              Gren        8      A
2021-10-01 00:00:00     Broad           Barb        5      A
2021-10-01 00:00:00     Voice           Jamai       23     A
2021-10-01 00:00:00     Voice           Trin        6      A  
2021-10-01 00:00:00     TV              Cur         7      A
2021-10-02 00:00:00     Broad           Jamai       2      A
2021-10-03 00:00:00     Broad           Trin        8      A
2021-10-04 00:00:00     Broad           Lucia       3      A
2021-10-04 00:00:00     TV              Anti        1      A
2021-10-04 00:00:00     Broad           Cur         8      A
2021-10-04 00:00:00     Voice           Barb        0      A
2021-10-04 00:00:00     TV              Vin         5      A
2021-10-04 00:00:00     Voice           Vin         1      A
2021-10-05 00:00:00     NAN             NAN         NAN    NAN
2021-10-06 00:00:00     NAN             NAN         NAN    NAN
2021-10-07 00:00:00     NAN             NAN         NAN    NAN  
2021-10-08 00:00:00     NAN             NAN         NAN    NAN      
2021-10-09 00:00:00     NAN             NAN         NAN    NAN  
2021-10-10 00:00:00     NAN             NAN         NAN    NAN  
2021-10-11 00:00:00     NAN             NAN         NAN    NAN  
2021-10-12 00:00:00     NAN             NAN         NAN    NAN      
2021-10-13 00:00:00     NAN             NAN         NAN    NAN      
2021-10-14 00:00:00     NAN             NAN         NAN    NAN      
2021-10-15 00:00:00     NAN             NAN         NAN    NAN      
...............
..............................etc

SAMPLE 结果格式：

week_start   week_end   product   country   name value(**average of values for each week**)

我尝试使用 groupby，但无法获取每个产品和国家/地区的周开始和结束时间。

此外，值应该是特定一周的平均值（加法）。关于如何实现这一目标的任何帮助？

Answer 1

第一步是为每行日期的开始日期和结束日期创建一个新列。这可以通过使用 offsets.Week:

来完成

import pandas as pd    

df['start'] = df['dates'] - pd.offsets.Week(weekday=6)
df['end'] = df['start'] + pd.offsets.Week(weekday=5)

从那里您可以使用 groupby 按开始、结束、产品和国家列进行分组，并对值列使用均值聚合方法：

df.groupby(['start','end','product','country']).agg({'value': 'mean'}).reset_index()

Python：根据 pandas 数据框中的每日数据计算周开始和周结束？

Python: Calculate week start and week end from daily data in pandas dataframe?

python

dataframe

python-3.x

pandas

pandas-groupby