param_grid = {
'criterion': ['gini', 'entropy'],
'n_estimators' : [len(train_cols), 100, 200],
'max_depth': [2, 5, 10, 15],
'min_samples_split': [2, 5, 10, 15],
'min_samples_leaf': [1, 3, 5, 10, 15],
'class_weight': [None, 'balanced']
}

grid_search = GridSearchCV(RandomForestClassifier(), param_grid, scoring='f1', cv=5, n_jobs=-1)
data_filtered = halo_table.to_pandas()[train_cols]
data_balanced, target_balanced = balance_tables_and_array(data_filtered, halo_target)
X_train, X_test, y_train, y_test = train_test_split(data_balanced, target_balanced, test_size=0.5, random_state=(np.random.randint(0,200)))
grid_search.fit(X_train, y_train)
print(grid_search.best_params_)