Agriculture, energy, and water management are important sectors on which the development as well as economics of a nation depends. Rainfall is one of the critical factors to affect all these. The conventional method to predict rainfall in accurate manner due to their inability to capture complex atmosphere features and temporal dependencies. The present study offers a range of supervised Machine Learning (ML) algorithms to apply in predicting the occurrence and intensity of rainfall across the Australian continent. The proposed ML forecasting framework considers sixteen meteorological variables such as geographical location, temperature, humidity, atmospheric pressure, wind speed etc. taking into account the previous rainfall records. A comparative study is conducted among various ML models viz. K-Nearest Neighbors (KNN), Decision Tree, Linear Regression, Support Vector Machine (SVM), Long Short-Term Memory (LSTM), and Bidirectional Long Short-Term Memory (Bi-LSTM) networks based on the key performance metrics like accuracy, Jaccard score, F1 score, and Log Loss. The simulation results indicate that among these models, the simulation result of Bi-LSTM model is best in terms of forecasting accuracy of occurrence and intensity and computational efficiency. In addition, with achieving the computational efficiency, the proposed method has potential to overcome the issues of overfitting and data quality.