I have data in an array like so:
JavaScript
x
10
10
1
array([[ 5, 5, 5, 6, 9, 6, 6],2
[10, 4, 10, 3, 5, 3, 3],3
[10, 3, 10, 4, 5, 3, 4],4
[ 9, 6, 8, 8, 10, 6, 9],5
[10, 10, 10, 7, 10, 4, 4],6
[10, 6, 10, 5, 9, 7, 5],7
[ 9, 7, 10, 7, 10, 8, 10],8
[ 8, 5, 10, 7, 10, 7, 10],9
[ 7, 10, 10, 9, 10, 7, 8]])10
I want to sort it by the amount of non-10 values, and I also want to sort it in ascending order for rows, and in descending order of number of 10s:
JavaScript
1
3
1
arr = np.sort(arr, axis=1)2
arr = arr[(arr==10).sum(axis=1).argsort()][::-1]3
Output:
JavaScript
1
10
10
1
array([[ 4, 4, 7, 10, 10, 10, 10],2
[ 7, 7, 8, 9, 10, 10, 10],3
[ 5, 7, 7, 8, 10, 10, 10],4
[ 7, 7, 8, 9, 10, 10, 10],5
[ 5, 5, 6, 7, 9, 10, 10],6
[ 3, 3, 4, 4, 5, 10, 10],7
[ 3, 3, 3, 4, 5, 10, 10],8
[ 6, 6, 8, 8, 9, 9, 10],9
[ 5, 5, 5, 6, 6, 6, 9]])10
I want to implement a tie breaker system so that if the amount of 10s the same, it now orders by amount of 9s, then 8s, and so on. Expected output:
JavaScript
1
10
10
1
array([[ 4, 4, 7, 10, 10, 10, 10],2
[ 7, 7, 8, 9, 10, 10, 10],3
[ 7, 7, 8, 9, 10, 10, 10],4
[ 5, 7, 7, 8, 10, 10, 10],5
[ 5, 5, 6, 7, 9, 10, 10],6
[ 3, 3, 4, 4, 5, 10, 10],7
[ 3, 3, 3, 4, 5, 10, 10],8
[ 6, 6, 8, 8, 9, 9, 10],9
[ 5, 5, 5, 6, 6, 6, 9]])10
Advertisement
Answer
You can achieve it with numpy.frompyfunc.
The basic idea is to construct an array with the same rows, each element of which is a tuple containing the number of 10s, 9s, etc. Then apply numpy.argsort to this array and get the result.
JavaScript
1
32
32
1
import numpy as np2
3
arr = np.array([[ 5, 5, 5, 6, 9, 6, 6],4
[10, 4, 10, 3, 5, 3, 3],5
[10, 3, 10, 4, 5, 3, 4],6
[ 9, 6, 8, 8, 10, 6, 9],7
[10, 10, 10, 7, 10, 4, 4],8
[10, 6, 10, 5, 9, 7, 5],9
[ 9, 7, 10, 7, 10, 8, 10],10
[ 8, 5, 10, 7, 10, 7, 10],11
[ 7, 10, 10, 9, 10, 7, 8]])12
arr = np.sort(arr, 1)13
14
keys = sorted(set(arr.ravel()), reverse=True)15
16
def make_tuple(*argv):17
return tuple(argv)18
ufunc = np.frompyfunc(make_tuple, len(keys), 1)19
20
cnt_array = ufunc(*[(arr == k).sum(1) for k in keys])21
result = arr[cnt_array.argsort()[::-1]]22
print(result)23
# [[ 4 4 7 10 10 10 10]24
# [ 7 7 8 9 10 10 10]25
# [ 7 7 8 9 10 10 10]26
# [ 5 7 7 8 10 10 10]27
# [ 5 5 6 7 9 10 10]28
# [ 3 3 4 4 5 10 10]29
# [ 3 3 3 4 5 10 10]30
# [ 6 6 8 8 9 9 10]31
# [ 5 5 5 6 6 6 9]]32