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
    0.0277    0.0225    0.0223    0.0263    0.0223

b) mean
    0.0242

c) standard deviation
    0.0026

2) specificity ( 1 - false positives )

a) for all 5 runs
    0.9801    0.9845    0.9825    0.9783    0.9841

b) mean
    0.9819

c) standard deviation
     0.0027

3) sensitivity (true positives)

a) for all 5 runs
    0.9593    0.9660    0.9699    0.9662    0.9673

b) mean
    0.9657

c) standard deviation
    0.0039