Number of training examples:   10000
Number of validation examples:  5000 (for model selection)
Number of test examples:       10000

1) error rate

a) for all 5 runs
    0.0308    0.0306    0.0278    0.0367    0.0268

b) mean
    0.0305

c) standard deviation
    0.0039

2) specificity ( 1 - false positives )

a) for all 5 runs
    0.9742    0.9767    0.9665    0.9777    0.9798

b) mean
    0.9750

c) standard deviation
     0.0051

3) sensitivity (true positives)

a) for all 5 runs
    0.9609    0.9575    0.9814    0.9399    0.9626

b) mean
    0.9605

c) standard deviation
    0.0148