Table 1 Employed Metrics.
From: Generalising from conventional pipelines using deep learning in high-throughput screening workflows
\(Accuracy = TP/(TP + TN)\) | (Eq. 1) |
\(Precision = TP/(TP + FP)\) | (Eq. 2) |
\(Recall = TP/(TP + FN)\) | (Eq. 3) |
\(Specificity = TN/(TN+FP)\) | (Eq. 4) |
\(IoU = TP/ (TP + FP + FN)\) | (Eq. 5) |
\(BFS = 2 * Precision * Recall / (Recall + Precision)\) | (Eq. 6) |
