Python groupby进程csv

我有一份netowrk路由器使用情况的每日档案。我正在尝试查找每个唯一路由器（QIN）的uIN和uOUT列的最大值及其发生时间

我对熊猫和“群比”做了很多，但似乎没有得到我需要的最终结果

以下是数据示例：

Minute  QIN uIN uOUT
2/14/2018 16:00 Bundle-Ether1 on  (Router1.network.com) 0.10221 0.21195
2/14/2018 16:05 Bundle-Ether1 on  (Router1.network.com) 0.089865    0.18722
2/15/2018 16:10 Bundle-Ether1 on  (Router1.network.com) 0.07482 0.1705
2/16/2018 16:15 Bundle-Ether1 on  (Router1.network.com) 0.09176 0.18846
2/17/2018 16:20 Bundle-Ether1 on  (Router1.network.com) 0.11816 0.11785
2/14/2018 16:00 Bundle-Ether1 on  (Router2.network.com) 0.08786 0.15235
2/14/2018 16:05 Bundle-Ether1 on  (Router2.network.com) 0.07777 0.19253
2/15/2018 16:10 Bundle-Ether1 on  (Router2.network.com) 0.07552 0.14232
2/16/2018 16:15 Bundle-Ether1 on  (Router2.network.com) 0.1291  0.18758
2/17/2018 16:20 Bundle-Ether1 on  (Router2.network.com) 0.13361 0.11747

这是我的密码：

import pandas as pd

df = pd.read_csv('c://router_data.csv')
df['Minute'] = pd.todatetime(df['Minute'])

df.set_index('Minute').groupby('QIN')['uIN'].resample("M").max()

结果:

Bundle-Ether1 on  (Router2.network.com) 0.13361
Bundle-Ether1 on  (Router1.network.com) 0.11816

我需要的结果是：

2/17/2018 16:20 Bundle-Ether1 on  (Router2.network.com) 0.13361
2/17/2018 16:20 Bundle-Ether1 on  (Router1.network.com) 0.11816

---

我建议合并。如有必要，您可以删除“uOUT”

import pandas as pd

df = pd.read_csv('C:\\router.csv', parse_dates=['Minute'], index_col='Minute')
df1 = df.groupby('QIN')['uIN'].max().reset_index()
df1 = df1.merge(df.reset_index(), on=['QIN', 'uIN']).set_index(['Minute', 'QIN'])

Out[191]: 
                                                                uIN     uOUT
Minute              QIN                                                     
2018-02-17 16:20:00 Bundle-Ether1 on (Router1.network.com)  0.11816  0.11785
                    Bundle-Ether1 on (Router2.network.com)  0.13361  0.11747

---

是的，那会有用的。嗯，我想把这件事弄得比需要的更复杂。谢谢您。