Table 1 Details of the metrics used to evaluate the CNN models in this study.
From: Deep learning twined spatial analysis for detection of mysterious fairy circles
Evaluation metric | Formula |
|---|---|
Accuracy | \(\:Acc=\:\frac{No.\:of\:correct\:predictions}{Total\:no.\:of\:predictions}\) |
BCE loss | \(\:L\left(y,\:\widehat{y}\right)=\:-(ylog\left(\widehat{y}\right)+(1-y)\text{l}\text{o}\text{g}(1-\widehat{y})\) |
Cohen kappa | \(\:\kappa\:=\:\frac{{p}_{0}-\:{p}_{e}}{1-\:{p}_{e}}\) |
AUC | \(\:AUC=\:\sum\:_{i=1}^{n}({FPR}_{i+1}\:-\:{FPR}_{i})\left(\frac{{TPR}_{i+1}+{TPR}_{i}}{2}\right)\) |
Precision | \(\:precision=\:\frac{TP}{TP+FP}\) |
Recall | \(\:recall=\:\frac{TP}{TP+FP}\) |
F1 | \(\:F1=\:2\:\times\:\frac{precision\:\times\:recall}{precision+recall}\) |
