df = pandas.read_excel('test.xlsx')
df = pandas.pivot_table(df, index=['a'], columns=['b'], aggfunc=numpy.sum)
先用pivot table 把表变成这样
pv uv
20180105 1 1
20180106 1 1
20180107 1 1
20180108 1 1
20180109 1 1
20180110 1 1
20180111 1 1
20180112 2 2
20180113 2 2
20180114 2 2
20180115 2 2
20180116 2 2
20180117 2 2
20180118 2 2
想计算周日均数据变成这样
pv uv
20180105~20180111 1 1
20180111~20180118 2 2
https://jingyan.baidu.com/article/2d5afd69ef4fcf85a2e28ef6.html