TABLE 1.

Brief descriptions of 10 candidate regression models, including the related packages and their parameters (default settings).

Model	Package	Key hyperparameters
ETR	scikit-learn 0.23.2 (from sklearn.ensemble import ExtraTreesRegressor)	“n_estimators”: 100, “max_depth”: None, “min_samples_leaf”: 1, ‘”min_samples_split”: 2, “max_features: auto”
RFR	scikit-learn 0.23.2 (from sklearn.ensemble import RandomForestRegressor)	“n_estimators”: 100, “max_depth”: None, “min_samples_leaf”: 1, “min_samples_split”: 2, ‘”max_features”: auto
BR	scikit-learn 0.23.2 (from sklearn.ensemble import BaggingRegressor)	“n_estimators”: 10, “max_depth”: 1.0, “max_samples”: 1.0
GBR	scikit-learn 0.23.2 (from sklearn.ensemble import GradientBoostingRegressor)	“n_estimators”: 100, “max_depth”: 3, “min_samples_leaf”: 1, “min_samples_split”: 2, “alpha”: 0.9, “learning_rate”: 0.1, “max_features”: None
ABR	scikit-learn 0.23.2 (from sklearn.ensemble import AdaBoostRegressor)	“n_estimators”: 50, “loss”: linear, “learning_rate”: 1.0
XGBR	xgboost 1.3.3 (from xgboost import XGBRegressor)	“n_estimators”: 100, “max_depth”: None, “min_child_weight”: None, “gamma”: None, “colsample_bytree”: None, “subsample”: None, “reg_alpha”: None, “reg_lambda”: None, “learning_rate”: None
SVR	scikit-learn 0.23.2 (from sklearn.svm import SVR)	“C”: 1.0, “gamma”: scale, “epsilon”: 0.1, “kernel”: rbf
KNR	scikit-learn 0.23.2 (from sklearn.neighbors import KNeighborsRegressor)	“weights”: uniform, “n_neighbors”: 5, “p”: 2
DTR	scikit-learn 0.23.2 (from sklearn.tree import DecisionTreeRegressor)	“criterion”: squared_error, “max_depth”: None, “min_samples_leaf”: 1, “min_samples_split”: 2, “max_features”: None
MLR	scikit-learn 0.23.2 (from sklearn.linear_model import LinearRegression)	“fit_intercept”: True, “normalize”: deprecated, “n_jobs”: None, “copy_X”: True