F1 to F-beta

Model Evaluation

Image by Author

F1 Score

The F-1 score is a popular binary classification metric representing a balance between precision and recall. It is the Harmonic mean of precision and recall. The following equation can represent the F-1 score.

Image by Author

where Precision can be defined as the probability of positive predictions that are actual members of the positive class.

Image by Author

The recall is defined as the probability of the positive predictions among the actual …

beta data science deep learning evaluation-metric machine learning python