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
    0.0275    0.0231    0.0230    0.0265    0.0220

b) mean
    0.0244

c) standard deviation
    0.0024

2) specificity ( 1 - false positives )

a) for all 5 runs
    0.9772    0.9789    0.9804    0.9767    0.9840

b) mean
    0.9795

c) standard deviation
    0.0029

3) sensitivity (true positives)

a) for all 5 runs
    0.9646    0.9736    0.9714    0.9683    0.9683

b) mean
    0.9693

c) standard deviation
    0.0034