回答編集履歴
1
サンプルを追加
answer
CHANGED
|
@@ -23,4 +23,20 @@
|
|
|
23
23
|
#3 13.000000
|
|
24
24
|
#4 14.800000
|
|
25
25
|
#Name: x1,dtype: float64
|
|
26
|
+
```
|
|
27
|
+
|
|
28
|
+
---
|
|
29
|
+
**【追記】**
|
|
30
|
+
|
|
31
|
+
新しいフレームワークを作成するサンプル
|
|
32
|
+
|
|
33
|
+
```Python
|
|
34
|
+
df_new = df2.groupby((df2.x2 != 0).cumsum()).apply(lambda d:pd.Series([d.index[0], d.index[-1], d['x1'].mean()],index=['start','end','mean_x1']))
|
|
35
|
+
print(df_new)
|
|
36
|
+
# start end mean_x1
|
|
37
|
+
#x2
|
|
38
|
+
#1 2017-07-01 00:00:00 2017-07-01 05:00:00 13.433333
|
|
39
|
+
#2 2017-07-01 06:00:00 2017-07-01 10:00:00 13.860000
|
|
40
|
+
#3 2017-07-01 11:00:00 2017-07-01 15:00:00 13.000000
|
|
41
|
+
#4 2017-07-01 16:00:00 2017-07-01 20:00:00 14.800000
|
|
26
42
|
```
|