熊猫数据框调整

Question

我有一些数据，如下：

+-----+-------+-------+--------------------+
| Sys | Event | Code  | Duration           |
+-----+-------+-------+--------------------+
|     |     1 |    65 |             355.52 |
|     |     1 |    66 |              18.78 |
|     |     1 |    66 |             223.42 |
|     |     1 |    66 |             392.17 |
|     |     2 |    66 |             449.03 |
|     |     2 |    66 |             506.03 |
|     |     2 |    66 |              73.93 |
|     |     3 |    66 |             123.17 |
|     |     3 |    66 |              97.85 |
+-----+-------+-------+--------------------+

现在，对于每个Code，我想对所有Durations的{​​{1}}求和，依此类推，而与Event = 1无关。我该如何处理？

Answer 1

正如DYZ所说：

df.groupby(['Code', 'Event']).Duration.sum()

输出：

Code  Event
65    1         355.52
66    1         634.37
      2        1028.99
      3         221.02
Name: Duration, dtype: float64