Run number 1
Pos Training images
8 2 10 7 4 
Neg Training images
11 8 12 2 13 
Pos Testing images
3 11 6 12 9 5 13 1 
Neg Testing images
9 3 6 1 10 7 5 4 
Pos test odds ratios
1.190424e+17 1.190424e+17 1.759629e+10 1.190424e+17 6.765201e+06 1.190424e+17 1.190424e+17 1.190424e+17 8 correct out of 8
Neg test odds ratios
1.759629e+10 2.601000e+03 8.400367e-18 8.400367e-18 8.400367e-18 8.400367e-18 6.765201e+06 2.184935e-14 5 correct out of 8

Run number 2
Pos Training images
6 9 2 3 5 
Neg Training images
11 4 8 7 2 
Pos Testing images
1 13 10 12 7 4 8 11 
Neg Testing images
1 12 13 10 6 3 5 9 
Pos test odds ratios
4.576794e+13 4.576794e+13 1.759629e+10 1.190424e+17 4.576794e+13 6.765201e+06 1.478153e-07 4.576794e+13 7 correct out of 8
Neg test odds ratios
8.400367e-18 2.184935e-14 8.400367e-18 8.400367e-18 8.400367e-18 6.765201e+06 4.576794e+13 3.844675e-04 6 correct out of 8

Run number 3
Pos Training images
7 13 12 4 10 
Neg Training images
9 6 1 5 2 
Pos Testing images
8 5 3 11 1 9 6 2 
Neg Testing images
13 10 12 4 7 11 8 3 
Pos test odds ratios
4.576794e+13 1.759629e+10 1.190424e+17 4.576794e+13 1.190424e+17 4.576794e+13 4.576794e+13 4.576794e+13 8 correct out of 8
Neg test odds ratios
1.190424e+17 8.400367e-18 8.400367e-18 2.184935e-14 2.184935e-14 5.683017e-11 8.400367e-18 1.478153e-07 7 correct out of 8

Overall results
Classification threshold = 10
Overall positive accuracy = 9.583333e-01
Overall negative accuracy = 7.500000e-01