Reference: pd.Grouper
is as follows: The X axis is the date from 2014-7-24 to 2015-3-5, and the Y axis is the daily sales.
Goal: Turn sales aggregated by day into aggregated by week, month, and year.
Solution:
df_new=df_normal.groupby([pd.Grouper(key='日期',freq='W')])[['销量']].sum().reset_index()
df_new
The result is as follows: it can be seen that the weekly sales are added up
日期 销量
0 2014-08-03 5698.10
1 2014-08-10 20778.00
2 2014-08-17 19775.20
3 2014-08-24 18904.80
4 2014-08-31 20946.10
5 2014-09-07 21164.20
6 2014-09-14 18832.70
7 2014-09-21 18376.70
8 2014-09-28 24253.44
9 2014-10-05 20048.80
10 2014-10-12 19328.00
11 2014-10-19 17779.10
12 2014-10-26 15300.10
13 2014-11-02 15130.20
14 2014-11-09 15395.80
15 2014-11-16 19114.90
16 2014-11-23 17706.70
17 2014-11-30 17485.20
18 2014-12-07 15847.30
19 2014-12-14 19300.60
20 2014-12-21 13925.50
21 2014-12-28 15134.40
22 2015-01-04 20265.00
23 2015-01-11 10048.80
24 2015-01-18 18539.80
25 2015-01-25 18447.70
26 2015-02-01 17835.50
27 2015-02-08 19593.20
28 2015-02-15 14531.70
29 2015-02-22 23654.10
30 2015-03-01 17901.30
The results after aggregating by days:
