Table 4: Training and testing results of experiments.

Experiment No. Network Size (Number of Computing Units)Learning Constant (Phase I)Learning Constant (Phase II)Training ResultTesting Result (Accuracy)
Input LayerHidden LayerOutput LayerNumber of Elapsed Training CycleError

1 4 12 6 0.3 0.011969.72712×10−498.99%
21919.94289×10−4100%
31859.80695×10−498.99%
41739.71459×10−497.98%
51889.67369×10−498.99%