Reservoir permeability prediction method based on XGBoost

The invention provides a model training method and device and a shale gas field reservoir permeability prediction method and device. The model training method comprises the following steps: dividing a training well and a test well from a plurality of wells of a target block; acquiring logging curve data corresponding to each well in the training well and the test well, preprocessing the logging curve data corresponding to each well, and taking the preprocessed logging curve data as characteristic parameters; acquiring the permeability corresponding to each well in the training well and the test well as a target parameter; constructing an XGBoost integrated model based on Boosting increment, training the XGBoost integrated model by adopting the characteristic parameters and the target parameters corresponding to the training wells, verifying the constructed XGBoost integrated model, and determining a target model; and evaluating the target model by adopting the characteristic parameters corresponding to the te