Skip to main content

Table 3 Number of parameters and obtained AUC for each architecture

From: CNNs for automatic glaucoma assessment using fundus images: an extensive validation

Model name # parameters (in millions) AUC
VGG16 138 0.9632 (0.0149)
VGG19 144 0.9686 (0.0158)
InceptionV3 23 0.9653 (0.0135)
ResNet50 25 0.9614 (0.0171)
Xception 22 0.9605 (0.0170)
  1. The best architecture in terms of AUC and number of parameters (in italic)