Use TPE to tune the LightGBM model, iterating 1000 times to obtain the optimal model and save the related parameters.
Based on SHAP analysis, plot the resulting model.
Using the indicator data of all regional points, predict their groundwater quality with the resulting model.
This is the dataset used to predict the groundwater quality of the entire region. It should be noted that the study area is large, and the original dataset is about 1 GB, making it difficult to upload. Therefore, only the first ten rows of data are shown.