Research Article

Comparing the Selected Transfer Functions and Local Optimization Methods for Neural Network Flood Runoff Forecast

Table 7

The MAE, NS, and R4M4E—trainings algorithms.


4-3-1
 FLET0.0170.0180.0160.980.890.970.1310.350.14
 HEST0.0150.0200.0170.980.900.970.1420.340.14
 PER0.0120.0170.0150.980.900.980.1240.340.12
 POL0.0160.0170.0150.970.880.970.1490.350.13
 LM0.0100.0150.0150.990.910.970.0930.330.12
 BP0.0140.0170.0160.980.890.970.1330.350.13
 BP_regul0.0150.0140.0150.980.890.970.1290.350.11

4-4-1
 FLET0.01720.0180.0160.980.890.970.140.350.13
 HEST0.01540.0200.0160.980.880.970.150.350.14
 PER0.01250.0170.0150.980.910.970.130.340.13
 POL0.01380.0190.0150.980.890.970.150.350.13
 LM0.00770.0140.0140.990.910.970.090.330.12
 BP0.01400.0180.0160.980.900.970.130.350.13
 BP_regul0.01450.0160.0150.980.900.970.120.340.12

4-5-1
 FLET0.01210.0160.0160.990.890.970.1170.350.13
 HEST0.01590.0200.0170.980.880.970.1370.350.12
 PER0.01290.0150.0150.990.900.980.1150.340.12
 POL0.01590.0220.0180.980.880.970.1430.350.14
 LM0.00720.0150.0130.990.910.980.0860.330.12
 BP0.01390.0160.0160.980.900.970.1220.350.13
 BP_regul0.01320.0140.0140.980.910.970.1320.340.12

4-6-1
 FLET0.01500.0160.0160.980.900.970.1320.350.13
 HEST0.01390.0180.0170.980.890.970.1340.350.14
 PER0.01170.0160.0140.990.910.970.1040.340.13
 POL0.01620.0180.0180.980.890.970.1360.350.14
 LM0.00760.0150.0140.990.910.970.0870.320.14
 BP0.01390.0170.0150.980.900.970.1310.350.14
 BP_regul0.01300.0150.0140.980.900.970.1300.340.12