Diabetic retinopathy (DR) is a complication of diabetes that causes retinal changes that affect vision. If not caught early, it can lead to blindness. Unfortunately, DR is not a reversible process where treatment only controls vision. Early detection and treatment of DR can reduce the risk of blindness. Compared with a computer-aided diagnosis system, the process of manually diagnosing retinal fundus DR images by ophthalmologists is time-consuming, laborious, expensive, and prone to misdiagnosis. Deep learning has become one of the most widely used methods in recent years and achieved great success in many fields, especially in the analysis and classification of medical imaging. The most popular and efficient deep learning technique for medical image processing used is a convolutional neural network. In this project, DR color fundus photos that were examined and analyzed were used to classify and recognize objects using deep learning techniques. Diabetic retinopathy was classified using a deep learning algorithm. The prediction of diabetic retinopathy is the main topic of this essay. The CNN model is trained using training datasets, and CNN will provide the likelihood of a diabetic eye infection.