This project applies machine learning to simulate and predict runoff from synthetic rainfall and hydrological data, enabling rainfall–runoff modeling without real datasets. It demonstrates data generation, model training, evaluation, and visualization to support hydrological forecasting and water resource management research.