Table 4 Performance of the H2O models for subsets of the data.

From: Near future flash flood prediction in an arid region under climate change

Metrics

Mean

SD

CV-1

CV-2

CV-3

CV-4

CV-5

MAE

0.3626

0.0216

0.3502

0.3965

0.3476

0.3720

0.3464

Mean residual deviance

5.1756

2.8326

2.8858

5.2562

4.9687

9.8394

2.9280

MSE

5.1756

2.8326

2.8858

5.2562

4.9687

9.8394

2.9280

Null deviance

6124.01

3398.01

3518.7

6396.25

5790.11

11679.8

3235.17

R2

0.0039

0.0016

0.0058

0.0034

0.0035

0.0017

0.0051

Residual deviance

6102.91

3397.51

3497.68

6370.56

5768.67

11659.8

3217.9

RMSE

2.2136

0.5865

1.6978

2.2926

2.2290

3.1367

1.7111

RMSLE

0.3221

0.0153

0.3255

0.3454

0.3111

0.3058

0.3229

  1. *CV-k and SD represent cross-validation for kth fold and standard deviation, respectively.
Back to article page