https://campus.datacamp.com/courses/ensemble-methods-in-python/combining-multiple-models?ex=7
pred_vote = clf_vote.predict(X_test)
score_vote = f1_score(y_test, pred_vote)
print('F1-Score: {:.3f}'.format(score_vote))
report = classification_report(y_test, pred_vote)
print(report)
Evaluating your ensemble
In the previous exercise, you built your first voting classifier. Let's now evaluate it and compare it to that of the individual models.
The individual models (clf_knn
, clf_dt
, and clf_lr
) and the voting classifier (clf_vote
) have already been loaded and trained.
Remember to use f1_score()
to evaluate the performance. In addition, you'll create a classification report on the test set (X_test
, y_test
) using the classifiation_report()
function.
Will your voting classifier beat the 58% F1-score of the decision tree?
- Use the voting classifier,
clf_vote
, to predict the labels of the test set,X_test
. - Calculate the F1-Score of the voting classifier.
- Calculate the classification report of the voting classifier by passing in
y_test
andpred_vote
toclassification_report()