Results from goodness-of-fit test for AI models at Level 1 and Level 2
. | RMSE . | R2 . | NSC . | ||||||
---|---|---|---|---|---|---|---|---|---|
Total . | Training . | Testing . | Total . | Training . | Testing . | Total . | Training . | Testing . | |
SFL at Level 1 | 3.41 | 1.14 | 7.3 | 0.930 | 0.990 | 0.869 | 0.909 | 0.990 | 0.348 |
NF at Level 1 | 2.8 | 0.81 | 6.07 | 0.940 | 0.995 | 0.913 | 0.938 | 0.995 | 0.540 |
SVM at Level 2 (AIMM) | 1.71 | 0.38 | 3.74 | 0.980 | 0.999 | 0.937 | 0.997 | 0.999 | 0.823 |
Percentage (%) of improvement (SFL-SVM) | 99 | 200 | 95 | 5 | 1 | 7 | 9 | 1 | 58 |
. | RMSE . | R2 . | NSC . | ||||||
---|---|---|---|---|---|---|---|---|---|
Total . | Training . | Testing . | Total . | Training . | Testing . | Total . | Training . | Testing . | |
SFL at Level 1 | 3.41 | 1.14 | 7.3 | 0.930 | 0.990 | 0.869 | 0.909 | 0.990 | 0.348 |
NF at Level 1 | 2.8 | 0.81 | 6.07 | 0.940 | 0.995 | 0.913 | 0.938 | 0.995 | 0.540 |
SVM at Level 2 (AIMM) | 1.71 | 0.38 | 3.74 | 0.980 | 0.999 | 0.937 | 0.997 | 0.999 | 0.823 |
Percentage (%) of improvement (SFL-SVM) | 99 | 200 | 95 | 5 | 1 | 7 | 9 | 1 | 58 |
The SVM parameters are: γ= 2, σ= 0.1.