Finding the probability of rain over a specific area is the aim of rainfall prediction. It is viewed as crucial for many industries, including agriculture. In order to support healthy populations, forecasting daily rainfall increases agricultural productivity and provides a steady supply of food and water. Data mining and machine learning are two methods that have been applied in a number of the research to predict rainfall using environmental datasets from different countries. Several hydrological models include rainfall as one of its components since it is crucial to the symmetry of the water cycle. One of the most challenging and unpredictable tasks with a big influence on society. Proactively reducing human and financial loss can be aided by timely and accurate forecasting. This study details a series of experiments that were conducted using well-known machine learning, deep learning, and ensemble models in order to construct models that are able to predict whether or not it would rain the next day based on the meteorological data for the previous day in significant Australian cities. In our first experiment, XGBoost, CatBoost, and Random Forest have done better. However if speed is a key factor, we can continue with Random Forest rather than XGBoost or CatBoost. The LGBMClassifier and XGBClassifier seem to do the best on our second experiment.