| Metric | Formula | Interpretation |
|---|---|---|
| Accuracy | $ \frac{TP+TN}{TP+TN+FP+FN} $ | Overall performance of model |
| Precision | $ \frac{TP}{TP+FN} $ | How accurate the positive predictions are |
| Recall Sensitivity | $ \frac{TP}{TP+FP} $ | Coverage of actual positive sample |
| Specificity | $ \frac{TN}{TN+FN} $ | Coverage of actual negative sample |
| F1 score | $ \frac{2(PrecisionRecall)}{2*Precision+Recall} $ | Hybrid metric useful for unbalanced classes |
| F-beta score | $ \frac{(1+\beta2)(precision*recall)}{\beta2*prediction+recall} $ | F-1 score generalized form (\beta = 1) |
Machine Learning in Python
发布时间 2023-12-18 09:53:37作者: ForHHeart