[英]How to group the cumulative sum of rain values into a new column for given timestamps
我有一个时间序列 dataframe 每个给定小时的降雨值。
这是当前的 dataframe：
print(assomption_rain_df.head(25))
date time rain_gauge_value
0 2020-08-05 00:00:00 5.3
1 2020-08-05 01:00:00 2.7
2 2020-08-05 02:00:00 2.2
3 2020-08-05 03:00:00 3.0
4 2020-08-05 04:00:00 1.5
5 2020-08-05 05:00:00 0.0
6 2020-08-05 06:00:00 0.0
7 2020-08-05 07:00:00 0.0
8 2020-08-05 08:00:00 0.1
9 2020-08-05 09:00:00 0.1
10 2020-08-05 10:00:00 0.1
11 2020-08-05 11:00:00 0.4
12 2020-08-05 12:00:00 0.5
13 2020-08-05 13:00:00 0.2
14 2020-08-05 14:00:00 0.3
15 2020-08-05 15:00:00 0.4
16 2020-08-05 16:00:00 0.2
17 2020-08-05 17:00:00 0.0
18 2020-08-05 18:00:00 0.0
19 2020-08-05 19:00:00 0.0
20 2020-08-05 20:00:00 0.0
21 2020-08-05 21:00:00 0.0
22 2020-08-05 22:00:00 0.0
23 2020-08-05 23:00:00 0.0
24 2020-08-06 00:00:00 0.0
这是我想要的 dataframe：
date time rain_gauge_value rain_gauge_value_6hr
0 2020-08-05 00:00:00 5.3 NaN
1 2020-08-05 01:00:00 2.7 NaN
2 2020-08-05 02:00:00 2.2 NaN
3 2020-08-05 03:00:00 3.0 NaN
4 2020-08-05 04:00:00 1.5 NaN
5 2020-08-05 05:00:00 0.0 14.7
6 2020-08-05 06:00:00 0.0 9.4
7 2020-08-05 07:00:00 0.0 6.7
8 2020-08-05 08:00:00 0.1 4.6
9 2020-08-05 09:00:00 0.1 1.7
10 2020-08-05 10:00:00 0.1 0.3
11 2020-08-05 11:00:00 0.4 0.7
12 2020-08-05 12:00:00 0.5 1.2
13 2020-08-05 13:00:00 0.2 1.4
14 2020-08-05 14:00:00 0.3 1.6
15 2020-08-05 15:00:00 0.4 1.9
16 2020-08-05 16:00:00 0.2 2.0
17 2020-08-05 17:00:00 0.0 1.6
18 2020-08-05 18:00:00 0.0 1.1
19 2020-08-05 19:00:00 0.0 0.9
20 2020-08-05 20:00:00 0.0 0.6
21 2020-08-05 21:00:00 0.0 0.2
22 2020-08-05 22:00:00 0.0 0.0
23 2020-08-05 23:00:00 0.0 0.0
24 2020-08-06 00:00:00 0.0 0.0
最终，我希望创建列rain_gauge_value_6hr, rain_gauge_value_12hr, rain_gauge_value_24hr, rain_gauge_value_36hr, rain_gauge_value_48hr
我读过一些类似的帖子，其中包括pandas.Grouper或pandas.DataFrame.resample但没有一个真正适用于我的情况。 我该如何实施？
编辑：
首先我试过：
assomption_rain_df['rain_gauge_value_6hr'] = assomption_rain_df['rain_gauge_value'].rolling(6).sum()
但它返回一个带有尴尬值的 dataframe ：
print(assomption_rain_df.head(25))
date time rain_gauge_value rain_gauge_value_6hr
0 2020-08-05 00:00:00 5.3 NaN
1 2020-08-05 01:00:00 2.7 NaN
2 2020-08-05 02:00:00 2.2 NaN
3 2020-08-05 03:00:00 3.0 NaN
4 2020-08-05 04:00:00 1.5 NaN
5 2020-08-05 05:00:00 0.0 1.470000e+01
6 2020-08-05 06:00:00 0.0 9.400000e+00
7 2020-08-05 07:00:00 0.0 6.700000e+00
8 2020-08-05 08:00:00 0.1 4.600000e+00
9 2020-08-05 09:00:00 0.1 1.700000e+00
10 2020-08-05 10:00:00 0.1 3.000000e-01
11 2020-08-05 11:00:00 0.4 7.000000e-01
12 2020-08-05 12:00:00 0.5 1.200000e+00
13 2020-08-05 13:00:00 0.2 1.400000e+00
14 2020-08-05 14:00:00 0.3 1.600000e+00
15 2020-08-05 15:00:00 0.4 1.900000e+00
16 2020-08-05 16:00:00 0.2 2.000000e+00
17 2020-08-05 17:00:00 0.0 1.600000e+00
18 2020-08-05 18:00:00 0.0 1.100000e+00
19 2020-08-05 19:00:00 0.0 9.000000e-01
20 2020-08-05 20:00:00 0.0 6.000000e-01
21 2020-08-05 21:00:00 0.0 2.000000e-01
22 2020-08-05 22:00:00 0.0 -2.609024e-15
23 2020-08-05 23:00:00 0.0 -2.609024e-15
24 2020-08-06 00:00:00 0.0 -2.609024e-15
您正在寻找DataFrame.rolling 。 它创建一个大小为n的滚动 window，您可以使用它执行操作。
你要
df["rain_gauge_value_6hr"] = df["rain_gauge_value"].rolling(6).sum()
当pandas中的某一列为0时如何重置每组的累积总和
[英]How to reset cumulative sum per group when a certain column is 0 in pandas
如何创建两个具有基于列值的累加总和并具有分组依据的列
[英]How to create two columns having cumulative sum based on a column value and having group by
从累积值计算非累积到新列（Python）
[英]Calculating Non-Cumulative into a new column from cumulative values (Python)
如何基于具有相同值的另一列将累积总和应用于 numpy 列？
[英]How do I apply cumulative sum to a numpy column based on another column having the same values?
声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.