Machine Learning in Python
发布时间 2023-12-18 09:53:37作者: ForHHeart
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) |