Compute precision, recall and F1 from true positives, false positives and false negatives. Useful for threshold selection on imbalanced data.
精确率关注预测正确性,召回率关注覆盖度。
| Item | Formula / Note | Source / Standard |
|---|---|---|
| Precision | precision is true positives over true plus false positives | ISO/IEC 22989:2022 AI concepts |
| Recall | recall is true positives over true positives plus false negatives | ISO/IEC 22989:2022 AI concepts |
| F1 score | F1 is the harmonic mean of precision and recall with equal weighting | van Rijsbergen 1979, Information Retrieval |