F1 score

As you've discovered, there's a tradeoff between precision and recall. Both are important metrics, and depending on how the business is trying to model churn, you may want to focus on optimizing one over the other. Often, stakeholders are interested in a single metric that can quantify model performance. The AUC is one metric you can use in these cases, and another is the F1 score, which is calculated as below:

2 * (precision * recall) / (precision + recall)

The advantage of the F1 score is it incorporates both precision and recall into a single metric, and a high F1 score is a sign of a well-performing model, even in situations where you might have imbalanced classes. In scikit-learn, you can compute the f-1 score using using the f1_score function.

Import f1_score from sklearn.metrics.
Print the F1 score of the trained random forest.

Exercise

F1 score

Instructions