Figure 2
From: Application of image processing and transfer learning for the detection of rust disease
The effect of different learning rates (0.01, 0.001, and 0.0001) on the accuracy of the residual networks (ResNet), Xception, EfficientNetB4, and MobileNetV2 pre-trained conventional neural network (CNN) models using different epoch numbers where Adaptive Moment Estimation (Adam) was used as the optimizer of the models.
