Table 5 Quantity of reinforcement and concrete in each project.
Model | Hyper-parameters | Value |
|---|---|---|
BP | Number of iterations | 2000 |
Learning rate | 0.01 | |
Number of hidden layer nodes | 11 | |
Minimum error of training objective | 0.00001 | |
Learning function | trainlm | |
RF | Estimator | 100 |
Max depth | 5 | |
XGBoost | Estimator | 500 |
Max depth | 3 | |
Learning rate | 0.01 | |
SSA | Early warning value | 0.6 |
Population size | 30 | |
Maximum number of iterations | 100 | |
Upper boundary of weight threshold | 5 | |
Weight threshold lower boundary | −5 | |
Proportion of searchers | 0.7 | |
Proportion of early warning personnel | 0.2 |
