-
Notifications
You must be signed in to change notification settings - Fork 11
/
Copy pathtiming
83 lines (58 loc) · 2.27 KB
/
timing
1
2
3
4
5
6
7
8
9
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
solver_test.py
Ubuntu 16.10 VirtualBox, 4 cores, MacBook Pro 2.3GHz Intel i7, gperftools 2.5, oof2 2.1.13, gcc 6.2.0
OpenMP tcmalloc user real sys
Y Y 9:16 8:43 0:06
Y Y 9:23 8:46 0:06
Y Y 9:32
Y Y 9:35
N Y 7:53 ??? openMP is slower?
N Y 7:49 8:20 0:03
N N 8:04
Y N 9:15
Debian (nestor), 8 cores, gcc 4.9.2, gperftools 2.2
N Y 12:51 13:45
N Y 12:50 13:40 0:24
Y Y 18:21 13:46 0:20
--------------------------------
3/10/23 Compare numpy and non-numpy versions, and the old 2.1.19
Run OOF.Image.Modify.Fade 30 times on a 476x444 rgb image
Fade is hand-coded using the ImageMagick API in non-numpy and 2.1.19
Non-numpy, with graphics display
real 0m33.557s user 0m31.041s sys 0m3.911s
real 0m33.236s user 0m31.086s sys 0m3.834s
Numpy, with graphics display
real 0m8.903s user 0m5.832s sys 0m3.781s
real 0m8.646s user 0m5.641s sys 0m3.775s
Version 2.1.19, with graphics display
real 0m26.611s user 0m26.334s sys 0m0.509s
real 0m26.106s user 0m25.817s sys 0m0.512s
Non-numpy, batch mode
real 0m41.009s user 0m27.516s sys 0m0.511s <-- outlier?
real 0m26.006s user 0m25.061s sys 0m0.490s
real 0m27.190s user 0m25.045s sys 0m0.489s
Numpy, batch mode
real 0m1.509s user 0m1.276s sys 0m0.534s
real 0m1.250s user 0m1.276s sys 0m0.537s
Version 2.1.19, batch mode
real 0m26.628s user 0m24.512s sys 0m0.260s
real 0m25.464s user 0m23.712s sys 0m0.246s
Run OOF.Image.Modify.Blur(radius=5, sigma=3) 30 times on the same image
Blur is built-in to numpy and ImageMagick
Non-numpy, with graphics display
real 0m9.567s user 0m6.843s sys 0m3.688s
real 0m9.154s user 0m6.535s sys 0m3.674s
Numpy, with graphics display
real 0m9.697s user 0m6.736s sys 0m3.795s
real 0m9.187s user 0m6.216s sys 0m3.781s
Version 2.1.19, with graphics
real 0m3.178s user 0m2.417s sys 0m0.564s
real 0m2.869s user 0m2.468s sys 0m0.464s
Non-numpy, batch mode
real 0m1.800s user 0m1.532s sys 0m0.463s
real 0m2.102s user 0m1.536s sys 0m0.460s
Numpy, batch mode
real 0m1.822s user 0m1.484s sys 0m0.533s
real 0m1.820s user 0m1.483s sys 0m0.535s
Version 2.1.19, batch mode
real 0m1.865s user 0m0.885s sys 0m0.232s
real 0m1.856s user 0m0.886s sys 0m0.224s