rf_different_ensembles_methods.py

# Build random forest classifier (same config)
rf_clf = RandomForestClassifier(n_estimators=500,
                                max_features=0.25,
                                criterion="entropy",
                                class_weight="balanced")

# Build model with no sampling
pip_orig = make_pipeline(Imputer(strategy="mean"),
                         RobustScaler(),
                         rf_clf)
scores = cross_val_score(pip_orig,
                         X_train, y_train,
                         scoring="roc_auc", cv=10)
print(f"Original model's average AUC: {scores.mean():.3f}")

# Build model with undersampling
pip_undersample = imb_make_pipeline(Imputer(strategy="mean"),
                                    RobustScaler(),
                                    RandomUnderSampler(),
                                    rf_clf)
scores = cross_val_score(pip_undersample,
                         X_train, y_train,
                         scoring="roc_auc", cv=10)
print(f"Under-sampled model's average AUC: {scores.mean():.3f}")

# Build model with oversampling
pip_oversample = imb_make_pipeline(Imputer(strategy="mean"),
                                   RobustScaler(),
                                   RandomOverSampler(),
                                   rf_clf)
scores = cross_val_score(pip_oversample,
                         X_train, y_train,
                         scoring="roc_auc", cv=10)
print(f"Over-sampled model's average AUC: {scores.mean():.3f}")

# Build model with EasyEnsemble
resampled_rf = BalancedBaggingClassifier(base_estimator=rf_clf,
                                         n_estimators=10,
                                         random_state=123)
pip_resampled = make_pipeline(Imputer(strategy="mean"),
                              RobustScaler(),
                              resampled_rf)
scores = cross_val_score(pip_resampled,
                         X_train, y_train,
                         scoring="roc_auc", cv=10)
print(f"EasyEnsemble model's average AUC: {scores.mean():.3f}")

# Build model with SMOTE
pip_smote = imb_make_pipeline(Imputer(strategy="mean"),
                              RobustScaler(),
                              SMOTE(),
                              rf_clf)
scores = cross_val_score(pip_smote,
                         X_train, y_train,
                         scoring="roc_auc", cv=10)
print(f"SMOTE model's average AUC: {scores.mean():.3f}")