Mathematical Problems in Engineering

Research Article

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

Table 9

The MAE, NS, and R4MS4E on activation functions, the best models are marked with bold font.
 
4-3-1          
 CL 0.10 0.10 0.10 0.72 0.52 0.68 0.40 0.63 0.42
 CLm 0.06 0.06 0.06 0.82 0.650.81 0.34 0.71 0.36
 HT 0.06 0.06 0.06 0.81 0.60 0.79 0.34 0.74 0.37
 LL 0.08 0.08 0.08 0.76 0.57 0.73 0.34 0.64 0.37
 RS 0.060.06 0.050.82 0.65 0.80 0.310.610.34
4-4-1          
 CL 0.10 0.10 0.10 0.71 0.53 0.68 0.40 0.62 0.41
 CLm 0.06 0.06 0.05 0.83 0.66 0.82 0.30 0.67 0.33
 HT 0.06 0.06 0.06 0.83 0.63 0.81 0.32 0.71 0.36
 LL 0.08 0.08 0.07 0.79 0.59 0.75 0.31 0.62 0.35
 RS 0.050.050.05 0.82 0.67 0.81 0.290.590.32
4-5-1          
 CL 0.10 0.10 0.10 0.72 0.56 0.68 0.41 0.62 0.42
 CLm 0.05 0.05 0.05 0.84 0.68 0.84 0.29 0.69 0.32
 HT 0.05 0.05 0.05 0.84 0.64 0.83 0.29 0.70 0.33
 LL 0.07 0.07 0.07 0.82 0.61 0.78 0.30 0.59 0.33
 LS 0.07 0.07 0.06 0.82 0.60 0.79 0.28 0.56 0.31
 RS 0.05 0.050.05 0.84 0.69 0.83 0.28 0.58 0.30
4-6-1          
 CL 0.10 0.10 0.10 0.75 0.59 0.69 0.42 0.60 0.42
 CLm 0.050.050.05 0.84 0.69 0.83 0.28 0.65 0.31
 HT 0.05 0.05 0.05 0.85 0.65 0.83 0.28 0.65 0.32
 LL 0.07 0.07 0.07 0.81 0.62 0.78 0.29 0.57 0.32
 LS 0.07 0.07 0.06 0.82 0.60 0.78 0.280.55 0.31
 RS 0.05 0.05 0.05 0.84 0.70 0.82 0.28 0.55 0.30