2a

main.py

# CITATION: https://scikit-learn.org/stable/modules/permutation_importance.html#permutation-importance
# CITATION: https://scikit-learn.org/stable/modules/generated/sklearn.inspection.permutation_importance.html
from sklearn.inspection import permutation_importance

models = {
    'Nearest Neighbors': knn,
    'Naive Bayes': nb,
    'Logistic Regression': log_reg,
    'Support Vector Machines': svm,
    'Decision Tree': dt,
    'Random Forest': rf
}

# Set input data for permutation importance
X_test_dict = {
    'Nearest Neighbors': X_test_scaled,
    'Naive Bayes': X_test,
    'Logistic Regression': X_test_scaled,
    'Support Vector Machines': X_test_scaled,
    'Decision Tree': X_test,
    'Random Forest': X_test
}

for name, model in models.items():
  # CITATION: https://scikit-learn.org/stable/modules/model_evaluation.html#scoring-parameter
  result = permutation_importance(
      model, X_test_dict[name], y_test,
      scoring='accuracy', n_repeats=30, random_state=1234
  )

  sorted_idx = result.importances_mean.argsort()[::-1][:5]
  print(f'\n{name} - Top 5 Features (in terms of impact on Accuracy):')
  for i in sorted_idx:
    print(f'{feature_names[i]:<25}: {result.importances_mean[i]:.4f} +/- {result.importances_std[i]:.4f}')