Skip to main content
. 2022 Mar 25;52(2):112–122. doi: 10.4041/kjod.2022.52.2.112

Table 4.

Precision, recall, and F1-score of the ResNet-101 model in the training, validation, and test sets in both the six-class and three-class classifications

Classification Class Training set Validation set Test set
Precision Recall F1-score Precision Recall F1-score Precision Recall F1-score
Six-class CS1 0.84 0.72 0.78 0.60 0.60 0.60 0.60 0.60 0.60
CS2 0.52 0.58 0.55 0.55 0.60 0.57 0.64 0.70 0.67
CS3 0.64 0.61 0.62 0.50 0.67 0.57 0.25 0.33 0.29
CS4 0.75 0.80 0.77 0.67 0.72 0.69 0.52 0.60 0.56
CS5 0.69 0.71 0.70 0.65 0.57 0.61 0.67 0.57 0.61
CS6 0.83 0.81 0.82 0.65 0.61 0.63 0.88 0.78 0.82
Three-class Class I 0.97 0.95 0.99 0.78 0.93 0.85 0.59 0.87 0.70
Class II 0.90 0.87 0.94 0.82 0.58 0.68 0.82 0.29 0.43
Class III 0.94 0.99 0.88 0.85 0.94 0.89 0.80 1.0 0.89

CS, cervical stage.