GoogleNet CNN Classifications for Diabetics Retinopathy

In the process of this work, an implementation of the CNN architecture that was powered by GoogleNet was created for the purpose of identifying cases of retinal diabetes. In order to categorize retinal pictures, the 144 layers of the GoogleNet CNN algorithm were utilized; these images were then trained utilizing a wide range of various training settings. Based on the findings of the comparison, the ADAM GoogleNet CNN option appears to be the most effective method for the diagnosis of diabetic retinal illness. It shows a 96% as accuracy in 3 min and 11 seconds with a good fit result.