Rainfall forecasting is one of the most critical management methods associated with water levels, agriculture, and disaster prevention. But it is not so simple to precisely forecast rainfall because the climatic condition is fluctuating along with other environmental conditions. There can be destruction and floods with the death of many people if the heavy rain is unforeseen. Thus, in this research study, the emphasis will be placed on determining the efficiency of ML and DL in accurately forecasting rainfall. Machine learning and deep learning models made the rainfalls’ predictions easier, yet the development stage. The techniques employed for rainfall prediction are Logistic regression, Decision tree, Random forest, SVM(Support vector machine), and CNN(Convolustion neural network). These models are validated using a test database and accuracy and ROC curve measures to find out the best model for rainfall prediction. This work attempts to compare some machine learning algorithms with deep learning algorithms and informs us which of the approach is optimal. As this study’s result indicates,the proposed CNN model in this paper is the most precise model compared to the others.