Hyper-parameter tuning using SMA
| Model | Hyperparameter chosen | |
|---|---|---|
| D1 dataset | D2 dataset | |
| AdaBoost | “n_estimators”: 50 “learning_rate”: 0.050 “max_depth”: 5 “Loss”: “exponential” | “n_estimators”: 50 “learning_rate”: 0.010 “max_depth”: 3 “Loss”: “exponential” |
| CatBoost | “iteration”: 147 “learning_rate”: 0.015 “depth”: 8 “l2_leaf_reg”: 3.521 “subsample”: 0.588 “random_strength”: 0.615 | “iteration”: 50 “learning_rate”: 0.028 “depth”: 8 “l2_leaf_reg”: 1.000 “subsample”: 0.500 “random_strength”: 0.037 |
| Extra tree | “n_estimators”: 200 “max_depth”: 6 “min_samples_split”: 14 “min_samples_leaf”: 3 “max_features”: 1.0 “Criterion”: “squared_error” | “n_estimators”: 73 “max_depth”: 6 “min_samples_split”: 14 “min_samples_leaf”: 3 “max_features”: 1.0 “Criterion”: “squared_error” |
| Gradient boosting | “n_estimators”: 147 “learning_rate”: 0.010 “max_depth”: 5 “min_samples_split”: 2 “min_samples_leaf”: 1 “subsample”: 0.500 “Loss”: “squared_error” | “n_estimators”: 50 “learning_rate”: 0.025 “max_depth”: 5 “min_samples_split”: 2 “min_samples_leaf”: 1 “subsample: 0.502 “Loss”: “squared_error” |
| Random forest | “n_estimators”: 473 “max_depth”: 15 “min_samples_split”: 20 “min_samples_leaf”: 2 “Max_features”: “sqrt” “Criterion”: “squared_error” “Bootstrap”: true | “n_estimators”: 96 “max_depth”: 6 “min_samples_split”: 12 “min_samples_leaf”: 10 “max_features”: 1.0 “Criterion”: “squared_error” “Bootstrap”: true |
| XGBoost | “learning_rate”: 0.106 “max_depth”: 8 “n_estimators”: 85 “gamma”: 51 “subsample”: 1.000 “colsample_bytree”: 0.823 | “learning_rate”: 0.169 “max_depth”: 10 “n_estimators”: 50 “gamma”: 152 “subsample”: 1.000 “colsample_bytree”: 0.500 |
| Model | Hyperparameter chosen | |
|---|---|---|
| D1 dataset | D2 dataset | |
| AdaBoost | “n_estimators”: 50 “learning_rate”: 0.050 “max_depth”: 5 “Loss”: “exponential” | “n_estimators”: 50 “learning_rate”: 0.010 “max_depth”: 3 “Loss”: “exponential” |
| CatBoost | “iteration”: 147 “learning_rate”: 0.015 “depth”: 8 “l2_leaf_reg”: 3.521 “subsample”: 0.588 “random_strength”: 0.615 | “iteration”: 50 “learning_rate”: 0.028 “depth”: 8 “l2_leaf_reg”: 1.000 “subsample”: 0.500 “random_strength”: 0.037 |
| Extra tree | “n_estimators”: 200 “max_depth”: 6 “min_samples_split”: 14 “min_samples_leaf”: 3 “max_features”: 1.0 “Criterion”: “squared_error” | “n_estimators”: 73 “max_depth”: 6 “min_samples_split”: 14 “min_samples_leaf”: 3 “max_features”: 1.0 “Criterion”: “squared_error” |
| Gradient boosting | “n_estimators”: 147 “learning_rate”: 0.010 “max_depth”: 5 “min_samples_split”: 2 “min_samples_leaf”: 1 “subsample”: 0.500 “Loss”: “squared_error” | “n_estimators”: 50 “learning_rate”: 0.025 “max_depth”: 5 “min_samples_split”: 2 “min_samples_leaf”: 1 “subsample: 0.502 “Loss”: “squared_error” |
| Random forest | “n_estimators”: 473 “max_depth”: 15 “min_samples_split”: 20 “min_samples_leaf”: 2 “Max_features”: “sqrt” “Criterion”: “squared_error” “Bootstrap”: true | “n_estimators”: 96 “max_depth”: 6 “min_samples_split”: 12 “min_samples_leaf”: 10 “max_features”: 1.0 “Criterion”: “squared_error” “Bootstrap”: true |
| XGBoost | “learning_rate”: 0.106 “max_depth”: 8 “n_estimators”: 85 “gamma”: 51 “subsample”: 1.000 “colsample_bytree”: 0.823 | “learning_rate”: 0.169 “max_depth”: 10 “n_estimators”: 50 “gamma”: 152 “subsample”: 1.000 “colsample_bytree”: 0.500 |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.