Table 3 The mean training and validation time for each deep learning model (in seconds)
From: Detection of developmental dysplasia of the hip in X-ray images using deep transfer learning
Data split | 70/30 | 90/10 |
|---|---|---|
Deep learning model | ||
SqueezeNet | 75.9 | 84.2 |
GoogLeNet | 129.8 | 152.8 |
Inceptionv3 | 356.7 | 413.5 |
DenseNet-201 | 1261.3 | 1451.6 |
MobileNetv2 | 586.7 | 674.6 |
ResNet-101 | 392.6 | 454.9 |
ResNet-50 | 185.7 | 205.8 |
ResNet-18 | 76.82 | 82.0 |
Xception | 1431.7 | 1629.0 |
Inception-ResNet-v2 | 1332.0 | 1363.7 |
ShuffleNet | 443.9 | 492.7 |
DarkNet-53 | 317.6 | 362.0 |
EfficientNet-b0 | 983.1 | 1331.5 |