Table 6.
Predictive performance of various machine-learning models for PNC.
| Models | Training set | Test set | ||||
|---|---|---|---|---|---|---|
| R2 | RMSE% | MAE% | R2 | RMSE% | MAE% | |
| RF | 0.98 | 0.06 | 0.04 | 0.97 | 0.08 | 0.06 |
| GBDT | 0.81 | 0.22 | 0.18 | 0.80 | 0.20 | 0.16 |
| XGB | 0.85 | 0.19 | 0.16 | 0.84 | 0.18 | 0.15 |
| SVR | 0.78 | 0.23 | 0.20 | 0.75 | 0.23 | 0.20 |
| KRR | 0.91 | 0.15 | 0.11 | 0.84 | 0.18 | 0.15 |
| GPR | 0.97 | 0.08 | 0.06 | 0.92 | 0.13 | 0.10 |