With a growing population, the demand for food increases every day. To meet the demands, the agricultural industry innovatively works on increasing production. A major setback is crop diseases affecting crop yield every year. Cassava is the staple diet of a major population of a country and when infected leads to famine and huge economic losses. This study aims to detect and classify the early onset of disease using pictures of the Cassava plant. An extensive dataset containing 21,397 images and divided into five classes, namely Cassava__bacterial_blight, Cassava__brown_streak_disease, Cassava__green_mottle, Cassava__healthy, and Cassava__mosaic_disease is used to train and test five CNN models and five transfer learning models. The models are compared based on six metrics to find the best model that can detect and classify the Cassava leaf disease. Xception, the transfer learning model, is found to be the best model with an accuracy of 88.73% making it the best classifier.