Dataset COIL100obj1

Basic characteristics COIL100obj1

48004

instances

7200

bags

32

features

72

positive bags

between 1 and 26

instances per bag

7128

negative bags

Results COIL100obj1

AUC dataset 201
classifier error
AUC EER meanPrec avCost
APR tau=0.999 NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
Diverse Density (10.000000) NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
EM-DD (10.000000) NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
MILBoost (100 rounds) 99.8 (0.1) 1.3 (0.1) 96.2 (2.6) 0.4 (0.1)
Citation MIL ref1, cite1 (h) 98.0 (0.0) 2.0 (0.0) 94.9 (0.0) 1.7 (0.0)
Citation MIL ref3, cite5 (h) NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
MI-SVM p=1.000000 NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
MI-SVM r=10.000000 NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
MILES p=1.000000 NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
MILES r=10.000000 NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
SVmil mahal.K NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
SVmil haussd.K NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
SVmil emd.K NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
Simple MIL with Logistic2, q=1 99.2 (0.0) 1.0 (0.1) 53.5 (1.4) 1.0 (0.0)
mean-inst+Logistic2 99.3 (0.9) 1.3 (1.2) 87.9 (2.7) 0.8 (0.7)
extremes+Logistic2 99.9 (0.1) 0.3 (0.6) 96.1 (1.0) 0.1 (0.2)
cov-coef+Logistic2 100.0 (0.0) 0.0 (0.0) 99.2 (0.7) 0.0 (0.0)
BagOfWords (k=10,diag)+Logistic2 96.6 (1.0) 7.8 (1.8) 36.2 (3.3) 3.6 (0.8)
BagOfWords (k=100,diag)+Logistic2 99.6 (0.2) 0.9 (0.5) 87.6 (2.6) 0.6 (0.2)
Pposterior MIL with N=30 99.7 (0.2) 1.0 (0.8) 88.6 (5.0) 0.6 (0.3)
Dissim.repr. AUC dataset 201
classifier error
AUC EER meanPrec avCost
minmin.K+Logistic2 100.0 (0.0) 0.0 (0.0) 99.2 (0.6) 0.0 (0.0)
summin.K+Logistic2 100.0 (0.1) 0.1 (0.1) 98.5 (3.4) 0.1 (0.1)
meanmin.K+Logistic2 100.0 (0.0) 0.0 (0.0) 100.0 (0.0) 0.0 (0.0)
meanmean.K+Logistic2 99.9 (0.0) 0.3 (0.0) 93.3 (0.7) 0.3 (0.0)
mahal.K+Logistic2 NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
haussd.K+Logistic2 NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
emd.K+Logistic2 NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)
linass.K+Logistic2 NaN (0.0) NaN (0.0) NaN (0.0) NaN (0.0)