Skip to main content

Table 2 Performance comparison with different sampling number of samples and training iterations

From: SMOTE-Boost-based sparse Bayesian model for flood prediction

Model

Sampling Number

RMSE

DC

FPE

Single week classifier

3000

99.26

0.79

256

 

4000

97.33

0.80

243

 

5000

96.43

0.80

251

Ensemble model with 3 iterations

3000

84.58

0.81

218

 

4000

80.27

0.82

207

 

5000

82.13

0.81

214

Ensemble model with 6 iterations

3000

75.27

0.82

200

 

4000

70.57

0.83

180

 

5000

73.96

0.82

196