Bananas are a fruit with promising economic value in Indonesia. They are an essential commodity for farmers, but diseases affecting banana plants can harm their livelihoods. Banana diseases initially attack the leaves, and in the early stages, they are difficult to differentiate with the naked eye due to farmers’ limited knowledge of pathogens. This research utilized the Convolutional Neural Network (CNN) method with transfer learning assistance using Google Colab to facilitate the classification of banana leaf diseases. The trained model experienced overfitting, so regularization was applied using dropout. The best model achieved an accuracy of 92%, precision of 92%, sensitivity of 91%, and an F1-score of 91% at a 70:20:10 ratio on epoch 80, as evaluated and validated using a confusion matrix. This study produced a reliable model for classifying banana leaf disease.