Sorting SQL, Python, Numpy, Pandas Comparison

	A	B	C	D	E	F	G	I
1	Language/ Library	Method	In-place or Return a Copy	Argument to Switch Sort Order to Descending	Type of Sort	Other Arguments	Notes	Default Axis
2	Python (vanilla)	my_list.sort()	In-place	reverse = True	timsort	key	For lists only.	n/a
3	Python (vanilla)	my_list = sorted(my_iterable)	Copy	reverse = True	timsort	key	For any iterable.	n/a
4	Numpy	my_ndarray.sort()	In-place	n/a - do outside with [::-1]	Default is quicksort. Alternative options: kind : ‘mergesort’, ‘heapsort’, ‘stable’.	axis=-1, kind='quicksort', order=None.	Quicksort is now introsort, which becomes heapsort if slow. Stable is mapped to mergesort. Mergesort doesn't use a mergesort. It uses timsort or radix sort under the hood, depening upon the data type.	last axis
5	Numpy	np.sort(my_array)	Copy	n/a - do outside with [::-1]	Default is quicksort Alternative options: kind : ‘mergesort’, ‘heapsort’, ‘stable’.	axis=-1, kind='quicksort', order=None.	Quicksort is now introsort, which becomes heapsort if slow. Stable is mapped to mergesort. Mergesort doesn't use a mergesort. It uses timsort or radix sort under the hood, depening upon the data type.	last axis
6	Pandas	df = df.sort_values(by='my_column')	Copy (unless inplace=True)	ascending = False	Default is numpy quicksort. Alternative options: kind : ‘mergesort’, ‘heapsort’, ‘stable’.	by, axis=0, inplace=False, kind='quicksort', na_position='last.	For DataFrame and Series. Uses Numpy for sorting under the hood.	last axis
7	TensorFlow	tf.sort(my_values)	Copy	direction = 'DESCENDING'	Finds the largest value in the tensor using top_k(). Uses CUB Cuda library, which wraps thrust, for parallel sort with GPU for large cases; algorithms vary - e.g. merge sort, radix sort.			last axis
8	PyTorch	torch.sort(my_values)	Copy	descending = True	Uses thrust CUDA library for parallel sort with GPU for large cases; algorithms vary - e.g. merge sort, radix sort.			last axis
9	SQL	ORDER BY my_column	Copy	DESC	Varies. Postgres uses a disk merge sort, quick sort, or heap sort depending upon the situation.			n/a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100