Research Article
Gradient Boosted Trees Predictive Models for Surface Roughness in High-Speed Milling in the Steel and Aluminum Metalworking Industry
Table 8
List of hyperparameters used for both machines with each model and general results.
| Experiment | Number of Trees (per class) | Maximal Depth | Minimum Rows | Accuracy on Grid Search Cross-Validation | Accuracy on Test Set |
| Slots (M1) | 40 | 4 | 5 | 82.51% +- 1.76% (82.50% micro average) | 78.00% | Slots (M2) | 60 | 6 | 4 | 89.67% +- 4.64% (89.62% micro average) | 61.54% | Geometries (M1) | 100 | 8 | 10 | 87.51% +- 2.93% (87.50% micro average) | 88.51% | Geometries (M2) | 40 | 4 | 8 | 89.57% +- 0.74% (89.58% micro average) | 85.71% |
|
|