Hyperparameter tuning with GridSearchCV and randomized search

                  from sklearn.model_selection import GridSearchCV, RandomizedSearchCV
from sklearn.ensemble import RandomForestClassifier

grid_search = GridSearchCV(
    estimator=RandomForestClassifier(random_state=42, n_jobs=-1),
    param_grid={
        'n_estimators': [200, 400],
        'max_depth': [None, 10, 20],
        'min_samples_leaf': [1, 3, 5],
    },
    scoring='roc_auc',
    cv=5,
    n_jobs=-1,
    verbose=1,
)

randomized_search = RandomizedSearchCV(
    estimator=RandomForestClassifier(random_state=42, n_jobs=-1),
    param_distributions={
        'n_estimators': [200, 300, 400, 500],
        'max_depth': [None, 8, 12, 16, 24],
        'min_samples_split': [2, 5, 10],
        'min_samples_leaf': [1, 2, 4],
    },
    n_iter=15,
    scoring='roc_auc',
    cv=5,
    n_jobs=-1,
    random_state=42,
)