The Effect of Adaptive Learning Rate on the Accuracy of Neural Networks

Learning rates in gradient descent algorithms have significant effects especially on the accuracy of a Capsule Neural Network (CNN). Choosing an appropriate learning rate is still issue to date. Many developers problem selecting for CNN leading low accuracies classification. This gap motivated this study assess effect developed There are no predefined and therefore it hard researchers know what will give good results. work, therefore, focused assessing by using different observing best performance. The contribution work CNNs improve during has assessed came up with most plant leaf disease Part images used was from PlantVillage dataset while others were Nepal database. pre-processed then subjected original model When 0.0001, performance 99.4% testing 100% training. 0.00001, highest 97% 99.9% lowest observed 81% 99% training when 0.001. that able achieve 0.0001. Convolutional 98%