Python 熊猫：如何对n天均匀采样的数据求和

我有一个熊猫数据框架，如下所示，这是一年内以15分钟的频率收集的数据。我想做的是添加一个天数据和下一天数据

我想做的是将第1天的数据与第2天的数据相加，依此类推n天，而不使用循环

例如：

第一天的数据是：

dttm_utc
2012-06-02 00:00:00    13.9678
2012-06-02 00:05:00    13.9678
2012-06-02 00:10:00    13.9678
2012-06-02 00:15:00    13.9678
2012-06-02 00:20:00     6.9839
2012-06-02 00:25:00    13.9678
2012-06-02 00:30:00    13.9678
2012-06-02 00:35:00    13.9678
2012-06-02 00:40:00    13.9678
2012-06-02 00:45:00    13.9678
2012-06-02 00:50:00     6.9839
2012-06-02 00:55:00    13.9678
2012-06-02 01:00:00    13.9678
2012-06-02 01:05:00    13.9678
2012-06-02 01:10:00     6.9839
2012-06-02 01:15:00    13.9678
2012-06-02 01:20:00    13.9678
2012-06-02 01:25:00    13.9678
2012-06-02 01:30:00     6.9839
2012-06-02 01:35:00    13.9678
2012-06-02 01:40:00    13.9678
2012-06-02 01:45:00    13.9678
2012-06-02 01:50:00     6.9839
2012-06-02 01:55:00    13.9678
2012-06-02 02:00:00    13.9678
2012-06-02 02:05:00    13.9678
2012-06-02 02:10:00     6.9839
2012-06-02 02:15:00    13.9678
2012-06-02 02:20:00    13.9678
2012-06-02 02:25:00    13.9678
                        ...   
2012-06-02 21:30:00    13.9678
2012-06-02 21:35:00    13.9678
2012-06-02 21:40:00    13.9678
2012-06-02 21:45:00     6.9839
2012-06-02 21:50:00    13.9678
2012-06-02 21:55:00     6.9839
2012-06-02 22:00:00    13.9678
2012-06-02 22:05:00    13.9678
2012-06-02 22:10:00    13.9678
2012-06-02 22:15:00     6.9839
2012-06-02 22:20:00    13.9678
2012-06-02 22:25:00     6.9839
2012-06-02 22:30:00    13.9678
2012-06-02 22:35:00    13.9678
2012-06-02 22:40:00    13.9678
2012-06-02 22:45:00     6.9839
2012-06-02 22:50:00    13.9678
2012-06-02 22:55:00    13.9678
2012-06-02 23:00:00    13.9678
2012-06-02 23:05:00     6.9839
2012-06-02 23:10:00    13.9678
2012-06-02 23:15:00    13.9678
2012-06-02 23:20:00     6.9839
2012-06-02 23:25:00    13.9678
2012-06-02 23:30:00    13.9678
2012-06-02 23:35:00    13.9678
2012-06-02 23:40:00    13.9678
2012-06-02 23:45:00     6.9839
2012-06-02 23:50:00    13.9678
2012-06-02 23:55:00    13.9678

同样，第2天的数据为：

2012-06-04 00:00:00     13.9678
2012-06-04 00:05:00      6.9839
2012-06-04 00:10:00     13.9678
2012-06-04 00:15:00     13.9678
2012-06-04 00:20:00      6.9839
2012-06-04 00:25:00     13.9678
2012-06-04 00:30:00     13.9678
2012-06-04 00:35:00     13.9678
2012-06-04 00:40:00     13.9678
2012-06-04 00:45:00      6.9839
2012-06-04 00:50:00     13.9678
2012-06-04 00:55:00     13.9678
2012-06-04 01:00:00     13.9678
2012-06-04 01:05:00     13.9678
2012-06-04 01:10:00      6.9839
2012-06-04 01:15:00     13.9678
2012-06-04 01:20:00     13.9678
2012-06-04 01:25:00     13.9678
2012-06-04 01:30:00      6.9839
2012-06-04 01:35:00     13.9678
2012-06-04 01:40:00     13.9678
2012-06-04 01:45:00     13.9678
2012-06-04 01:50:00      6.9839
2012-06-04 01:55:00     13.9678
2012-06-04 02:00:00     13.9678
2012-06-04 02:05:00     13.9678
2012-06-04 02:10:00      6.9839
2012-06-04 02:15:00     13.9678
2012-06-04 02:20:00     13.9678
2012-06-04 02:25:00     13.9678
                         ...   
2012-06-04 21:30:00    160.6302
2012-06-04 21:35:00    146.6623
2012-06-04 21:40:00    146.6623
2012-06-04 21:45:00    146.6623
2012-06-04 21:50:00    146.6623
2012-06-04 21:55:00    153.6462
2012-06-04 22:00:00    146.6623
2012-06-04 22:05:00    146.6623
2012-06-04 22:10:00    146.6623
2012-06-04 22:15:00    139.6784
2012-06-04 22:20:00    139.6784
2012-06-04 22:25:00    139.6784
2012-06-04 22:30:00    139.6784
2012-06-04 22:35:00    139.6784
2012-06-04 22:40:00    139.6784
2012-06-04 22:45:00    139.6784
2012-06-04 22:50:00    139.6784
2012-06-04 22:55:00    132.6945
2012-06-04 23:00:00    139.6784
2012-06-04 23:05:00    111.7427
2012-06-04 23:10:00    118.7266
2012-06-04 23:15:00    111.7427
2012-06-04 23:20:00    118.7266
2012-06-04 23:25:00    132.6945
2012-06-04 23:30:00    132.6945
2012-06-04 23:35:00    132.6945
2012-06-04 23:40:00    125.7106
2012-06-04 23:45:00    125.7106
2012-06-04 23:50:00    132.6945
2012-06-04 23:55:00    132.6945

看来你需要：

print (df.groupby(df.index.time)['dttm_utc'].sum())
00:00:00    27.9356
00:05:00    20.9517
00:10:00    27.9356
00:15:00    27.9356
00:20:00    13.9678
00:25:00    27.9356
00:30:00    27.9356
00:35:00    27.9356
00:40:00    27.9356
00:45:00    20.9517
00:50:00    20.9517
00:55:00    27.9356
01:00:00    27.9356
...
...

或：

---

你能不能简单地使用

sum（zip（day_1['value_column']，day_2['value_column']）

？

你的意思是“将第一天的值求和，然后将它们添加到第二天的值之和”？我的意思是将day_1[0]与day_2[0]相加，day_1[1]与day_2[1]相加，依此类推。抱歉，耶斯雷尔，我想把每个系列的值加起来。我的意思是将第1天[0]与第2天[0]相加，第1天[1]与第2天[1]相加，依此类推。直到天涯海角。jesrael：我的全部数据都在一个df中。一年的数据。我没有df2。我必须从df中提取每天的数据，并将其添加到第二天的数据中。我怎么做？对不起，我有点困惑。您能否获得具有所需输出的短路数据样本（3-4行）？

print (df.resample('D').sum())

            dttm_utc
2012-06-02  202.5331
2012-06-03       NaN
2012-06-04  230.4687