Mastering Machine Learning with scikit-learn - Second Edition

Multi-class classification performance metrics

As with binary classification, confusion matrices are useful for visualizing the types of errors made by the classifier. Precision, recall, and F1 score can also be computed for each of the classes and accuracy for all of the predictions can be calculated. Let's evaluate our classifier's predictions:

# In[8]:predictions = grid_search.predict(X_test)print('Accuracy: %s' % accuracy_score(y_test, predictions))print('Confusion Matrix:')print(confusion_matrix(y_test, predictions))print('Classification Report:')print(classification_report(y_test, predictions))# Out[8]:Accuracy: 0.636255286428Confusion Matrix:[[ 1124  1725   628    65    10] [  923  6049  6132   583    34] [  197  3131 32658  3640   137] [ 15 398 6530 8234 ...