Table 2 Model parameter configurations.
From: Applicability analysis of tree-based ensemble learning for air pollutant prediction models
Model name | Model object | Parameter initialization configuration | Parameter configuration |
|---|---|---|---|
RandomForest | RandomForestRegressor | max_depth = 10 | ‘n_estimators’: [100, 200] |
n_estimators = 200 | ‘max_depth’: [None, 10] | ||
max_features=’sqrt’ | ‘max_features’: [‘sqrt’, 0.33] | ||
n_jobs=-1 | – | ||
random_state = 42 | – | ||
GradientBoosting | GradientBoostingRegressor | n_estimators = 100 | ‘n_estimators’: [100, 200] |
learning_rate = 0.05 | ‘learning_rate’: [0.05, 0.1] | ||
max_depth = 3 | ‘max_depth’: [3, 5] | ||
subsample = 0.8 | – | ||
random_state = 42 | – | ||
DecisionTree | DecisionTreeRegressor | max_depth = 5 | ‘max_depth’: [5, 8, None] |
min_samples_split = 20 | ‘min_samples_split’: [10, 20] | ||
random_state = 42 | – |
