Skip to main content

Table 2 Proposed pre-trained classification models weight size and their top-1 accuracy performance on the ImageNet’s validation dataset

From: Single-cell conventional pap smear image classification using pre-trained deep neural network architectures

Model Size Top-1 Accuracy
NASNetLarge 343 MB 0.825
InceptionResNetV2 215 MB 0.803
Xception 88 MB 0.790
ResNet152V2 232 MB 0.780
InceptionV3 92 MB 0.779
DenseNet201 80 MB 0.773
ResNet101V2 171 MB 0.772
ResNet152 232 MB 0.766
ResNet101 171 MB 0.764
DenseNet169 57 MB 0.762