Study on Rainfall Prediction of Yibin City Based on GRU and XGBoost

rainfall is an important indicator of regional climate, which is closely related to people's production and life. Accurate rainfall prediction can help people make correct decisions and minimize disasters caused by rainfall as much as possible. In this paper, GRU and XGBoost algorithms are used to predict the rainfall in Yibin City, Sichuan Province, and the different prediction results of the two methods are compared. Experiments have found that the mean absolute error (MAE) of GRU and XGBoost models are 0.088 and 0.013 respectively; The mean square error (MSE) is 0.016 and 0.0003 respectively; The mean average absolute percentage error (MAPE) was 58.22 and 14.85 respectively. In the two prediction models, the prediction error of XGBoost model is smaller. In other words, the prediction accuracy of XGBoost model is higher. The XGBoost model has some advantages in processing rainfall and other time-series data. In future research, it can be applied to other fields for prediction research.