Number of training examples:     100
Number of validation examples:  5000 (for model selection)
Number of test examples:       10000

1) error rate

a) for all 5 runs
    0.1500    0.1116    0.0731    0.0721    0.0741

b) mean
    0.0962

c) standard deviation
    0.0344

2) specificity ( 1 - false positives )

a) for all 5 runs
    0.8399    0.8685    0.9385    0.9301    0.9301

b) mean

    0.9014

c) standard deviation
    0.0444

3) sensitivity (true positives)

a) for all 5 runs
    0.8662    0.9215    0.9082    0.9242    0.9189

b) mean
    0.9078

c) standard deviation
    0.0240