回答編集履歴
1
補足を追加
answer
CHANGED
|
@@ -1,3 +1,5 @@
|
|
|
1
|
+
`numpy.argsort`を使う。
|
|
2
|
+
|
|
1
3
|
```Python
|
|
2
4
|
import numpy as np
|
|
3
5
|
|
|
@@ -10,4 +12,50 @@
|
|
|
10
12
|
|
|
11
13
|
arr1 = arr[np.argsort(arr[:, 0])]
|
|
12
14
|
print(arr1)
|
|
15
|
+
```
|
|
16
|
+
|
|
17
|
+
```result
|
|
18
|
+
[[5 1]
|
|
19
|
+
[3 4]
|
|
20
|
+
[4 8]]
|
|
21
|
+
基準となる列のndarray: [5 3 4]
|
|
22
|
+
その列で昇順ソートした場合のインデックス: [1 2 0]
|
|
23
|
+
[[3 4]
|
|
24
|
+
[4 8]
|
|
25
|
+
[5 1]]
|
|
26
|
+
```
|
|
27
|
+
|
|
28
|
+
面倒くさいなら、いったんpandasに変換してソートしてからndarrayに戻す。
|
|
29
|
+
|
|
30
|
+
```Python
|
|
31
|
+
import numpy as np
|
|
32
|
+
import pandas as pd
|
|
33
|
+
|
|
34
|
+
arr = np.array([[5, 1], [3, 4], [4, 8]])
|
|
35
|
+
print(arr)
|
|
36
|
+
|
|
37
|
+
df = pd.DataFrame(arr)
|
|
38
|
+
print(df)
|
|
39
|
+
df.sort_values(0, inplace=True)
|
|
40
|
+
print(df)
|
|
41
|
+
|
|
42
|
+
arr1 = df.values
|
|
43
|
+
print(arr1)
|
|
44
|
+
```
|
|
45
|
+
|
|
46
|
+
```result
|
|
47
|
+
[[5 1]
|
|
48
|
+
[3 4]
|
|
49
|
+
[4 8]]
|
|
50
|
+
0 1
|
|
51
|
+
0 5 1
|
|
52
|
+
1 3 4
|
|
53
|
+
2 4 8
|
|
54
|
+
0 1
|
|
55
|
+
1 3 4
|
|
56
|
+
2 4 8
|
|
57
|
+
0 5 1
|
|
58
|
+
[[3 4]
|
|
59
|
+
[4 8]
|
|
60
|
+
[5 1]]
|
|
13
61
|
```
|