F-measure is considered as one of the better ways to assess the model. In lots of areas of data science, competition model performance is assessed using F-measure. It is a harmonic mean of precision and recall. The higher the value of the F1-score, the better the model is considered. F1-score provides equal weightage to precision and recall, which means it indicates a balance between both:
One drawback of F-measure is that it assigns equal weightage to precision and recall but in some examples, one needs to be higher than the other, which is the reason why the F1-score may not be an exact metric.
In the preceding sections, we have seen classification algorithms such as naive Bayes, decision trees, KNN, and SVMs. We have assessed the model performance using scikit-learn's accuracy_score() for model accuracy, precision_score() for model precision, recall_score() for model recall, and f1_score() for model F1-score.
We can also print the classification report to...
