Table 2 Performance parameters of pre-trained models VGG16, MobileNetV2 and ResNet101V2.

From: Accuracy of deep learning models in the detection of accessory ostium in coronal cone beam computed tomographic images

Performance (on 20 epochs)

VGG16

MobileNetV2

ResNet101V2

Training Accuracy

55

98

99

Valid Accuracy

51

66.67

81