@@ -31,42 +31,42 @@ Bottleneck comes with a benchmark suite::
31
31
32
32
>>> bn.bench()
33
33
Bottleneck performance benchmark
34
- Bottleneck 1.3.0.dev0; Numpy 1.11.3
34
+ Bottleneck 1.3.0.dev0; Numpy 1.12.1
35
35
Speed is NumPy time divided by Bottleneck time
36
36
NaN means approx one-fifth NaNs; float64 used
37
37
38
38
no NaN no NaN NaN no NaN NaN
39
39
(100,) (1000,1000)(1000,1000)(1000,1000)(1000,1000)
40
40
axis=0 axis=0 axis=0 axis=1 axis=1
41
- nansum 62 .3 1.6 2.0 2.3 2.5
42
- nanmean 230.7 2.4 2.4 3.5 2.9
43
- nanstd 260.2 2.1 2.2 2.7 2.6
44
- nanvar 250.5 2.1 2.3 2.8 2.6
45
- nanmin 43.3 0.7 1.9 0.8 2.6
46
- nanmax 45.9 0.7 2.1 1.0 3.3
47
- median 108.6 1.3 6.2 1.1 6.2
48
- nanmedian 108.8 5.7 6.6 5.5 6.6
49
- ss 27.2 1.2 1.2 1.6 1.6
50
- nanargmin 80.8 3.1 5.4 2.3 6.0
51
- nanargmax 95.0 3.2 5.4 2.3 6.0
52
- anynan 18.1 0.3 34.5 0.5 29.6
53
- allnan 39.7 146.3 126.9 117.0 96.3
54
- rankdata 56.1 2.5 2.5 2.8 2.9
55
- nanrankdata 60 .8 2.7 2.7 3.1 3.0
56
- partition 4.1 1.2 1.6 1.0 1.4
57
- argpartition 3.0 1.1 1.4 1.1 1.6
58
- replace 12.3 1.4 1.4 1.4 1.4
59
- push 3363.6 7.6 9.1 20.2 15.7
60
- move_sum 5046.7 67.5 147.9 192.9 211.7
61
- move_mean 12277.3 111.8 180.0 252.9 261.6
62
- move_std 10677.3 97.0 196.6 145.1 258.0
63
- move_var 13537.3 123.7 235.5 214.7 324.8
64
- move_min 2474.2 20.0 36.9 23.5 41.9
65
- move_max 2416.5 20.2 37.1 23.7 42.3
66
- move_argmin 3876.9 38.9 72.4 39.6 80.7
67
- move_argmax 3910.2 40.3 73.9 41.2 81.0
68
- move_median 2087 .3 148.2 161.9 148.4 160.7
69
- move_rank 1312 .5 1.8 2.1 2.3 2.7
41
+ nansum 67 .3 0.3 0.7 2.5 2.4
42
+ nanmean 194.8 1.9 2.1 3.4 3.1
43
+ nanstd 241.5 1.6 2.1 2.7 2.6
44
+ nanvar 229.7 1.7 2.1 2.7 2.5
45
+ nanmin 34.1 0.7 1.1 0.8 2.6
46
+ nanmax 45.6 0.7 2.7 1.0 3.7
47
+ median 111.0 1.3 5.6 1.0 4.8
48
+ nanmedian 108.8 5.9 6.7 5.6 6.7
49
+ ss 16.3 1.1 1.2 1.6 1.6
50
+ nanargmin 89.2 2.9 5.1 2.2 5.6
51
+ nanargmax 107.4 3.0 5.4 2.2 5.8
52
+ anynan 19.4 0.3 35.0 0.5 29.9
53
+ allnan 39.9 146.6 128.3 115.8 75.6
54
+ rankdata 55.0 2.6 2.3 2.9 2.8
55
+ nanrankdata 59 .8 2.8 2.2 3.2 2.5
56
+ partition 4.4 1.2 1.6 1.0 1.4
57
+ argpartition 3.5 1.1 1.4 1.1 1.6
58
+ replace 17.7 1.4 1.4 1.3 1.4
59
+ push 3440.0 7.8 9.5 20.0 15.5
60
+ move_sum 4743.0 75.7 156.1 195.4 211.1
61
+ move_mean 8760.9 116.2 167.4 252.1 258.8
62
+ move_std 8979.9 96.1 196.3 144.0 256.3
63
+ move_var 11216.8 127.3 243.9 225.9 321.4
64
+ move_min 2245.3 20.6 36.7 23.2 42.1
65
+ move_max 2223.7 20.5 37.2 24.1 42.4
66
+ move_argmin 3664.0 48.2 73.3 40.2 83.9
67
+ move_argmax 3916.9 42.0 75.4 41.5 81.2
68
+ move_median 2023 .3 166.8 173.7 153.8 154.3
69
+ move_rank 1208 .5 1.9 1.9 2.5 2.8
70
70
71
71
You can also run a detailed benchmark for a single function using, for
72
72
example, the command::
@@ -101,7 +101,7 @@ Install
101
101
Requirements:
102
102
103
103
======================== ====================================================
104
- Bottleneck Python 2.7, 3.4, 3.5; NumPy 1.11.3
104
+ Bottleneck Python 2.7, 3.4, 3.5; NumPy 1.12.1
105
105
Compile gcc, clang, MinGW or MSVC
106
106
Unit tests nose
107
107
======================== ====================================================
0 commit comments