Table 17 Models and hyperparameters used for SWLS and WHO-5 regression
From: Predicting subjective well-being in a high-risk sample of Russian mental health app users
Model | Hyperparameters |
|---|---|
AdaBoostRegressor | loss’: [‘linear’, ‘square’, ‘exponential’], ‘n_estimators’: [10,100] |
DecisionTreeRegressor | criterion’: [‘mae’], ‘max_depth’: [2,3], ‘min_samples_leaf’: [2], ‘max_leaf_nodes’: [3], ‘splitter’: [‘best’], ‘min_samples_split’: [2], ‘max_features’: [‘auto’] |
ElasticNet | alpha’: [100,10,1,0.1,0.01,0.001,0.0001], ‘normalize’: [False, True], ‘selection’: [‘cyclic’, ‘random’], ‘max_ iter’: [500,1000], ‘l1_ratio’: [0.25,0.5,0.75] |
Lasso | alpha’: [100,10,1,0.1,0.01,0.001,0.0001], ‘normalize’: [False, True], ‘selection’: [‘cyclic’, ‘random’],’max_iter’: [500,1000,2000] |
LinearRegression | normalize’: [False, True] |
RandomForestRegressor | n_estimators’: [2,5,10,20], ‘max_depth’: [2,3], ‘min_samples_split’: [2], ‘min_samples_leaf’: [1], ‘max_ features’: [‘auto’] |
Ridge | alpha’: [100,10,1,0.1,0.01,0.001,0.0001], ‘normalize’: [False, True] |