source: orange/Orange/testing/regression/results/orange/modules/statExamples.py.linux2.txt @ 9679:3879dea56188

Revision 9679:3879dea56188, 2.3 KB checked in by Miha Stajdohar <miha.stajdohar@…>, 2 years ago (diff)

Moved and renamed testing.

Line 
1
2method  CA  AP  Brier   IS
3bayes   0.903   0.902   0.175    0.759
4tree    0.846   0.845   0.286    0.641
5majrty  0.614   0.526   0.474   -0.000
6
7method  CA  AP  Brier   IS
8bayes   0.903+-0.019    0.902+-0.019    0.175+-0.036     0.759+-0.039
9tree    0.846+-0.016    0.845+-0.015    0.286+-0.030     0.641+-0.032
10majrty  0.614+-0.003    0.526+-0.001    0.474+-0.001    -0.000+-0.000
11
12Confusion matrix for naive Bayes:
13TP: 238, FP: 13, FN: 29.0, TN: 155
14
15Confusion matrix for naive Bayes:
16TP: 239, FP: 18, FN: 28.0, TN: 150
17
18Confusion matrix for naive Bayes for 'van':
19TP: 189, FP: 241, FN: 10.0, TN: 406
20
21Confusion matrix for naive Bayes for 'opel':
22TP: 86, FP: 112, FN: 126.0, TN: 522
23
24    bus van saab    opel
25bus 56  95  21  46
26van 6   189 4   0
27saab    3   75  73  66
28opel    4   71  51  86
29
30Sensitivity and specificity for 'voting'
31method  sens    spec
32bayes   0.891   0.923
33tree    0.816   0.893
34majrty  1.000   0.000
35
36Sensitivity and specificity for 'vehicle=van'
37method  sens    spec
38bayes   0.950   0.628
39tree    0.809   0.966
40majrty  0.000   1.000
41
42AUC (voting)
43     bayes: 0.974
44      tree: 0.930
45    majrty: 0.500
46
47AUC for vehicle using weighted single-out method
48bayes   tree    majority
490.783   0.800   0.500
50
51AUC for vehicle, using different methods
52                            bayes   tree    majority
53       by pairs, weighted:  0.789   0.870   0.500
54                 by pairs:  0.791   0.871   0.500
55    one vs. all, weighted:  0.783   0.800   0.500
56              one vs. all:  0.783   0.800   0.500
57
58AUC for detecting class 'van' in 'vehicle'
590.858   0.888   0.500
60
61AUCs for detecting various classes in 'vehicle'
62bus (218.000) vs others:    0.894   0.932   0.500
63van (199.000) vs others:    0.858   0.888   0.500
64saab (217.000) vs others:   0.699   0.687   0.500
65opel (212.000) vs others:   0.682   0.694   0.500
66
67    bus van saab
68van 0.933
69saab    0.820   0.828
70opel    0.822   0.825   0.519
71
72AUCs for detecting various pairs of classes in 'vehicle'
73van vs bus:     0.933   0.978   0.500
74saab vs bus:    0.820   0.938   0.500
75saab vs van:    0.828   0.879   0.500
76opel vs bus:    0.822   0.932   0.500
77opel vs van:    0.825   0.903   0.500
78opel vs saab:   0.519   0.599   0.500
79
80AUC and SE for voting
81bayes: 0.968+-0.015
82tree: 0.924+-0.022
83majrty: 0.500+-0.045
84
85Difference between naive Bayes and tree: 0.014+-0.062
86
87ROC (first 20 points) for bayes on 'voting'
881.000   1.000
890.970   1.000
900.910   1.000
910.881   1.000
920.821   1.000
930.806   1.000
940.791   1.000
950.761   1.000
960.746   1.000
970.731   1.000
980.701   1.000
990.687   1.000
1000.672   1.000
1010.672   0.991
1020.657   0.991
1030.642   0.991
1040.552   0.991
1050.537   0.991
1060.522   0.991
1070.507   0.991
Note: See TracBrowser for help on using the repository browser.