Geophysical oil and gas prediction method and device and storage medium

The invention provides a geophysical oil and gas prediction method and device and a storage medium. The method comprises the steps that logging data serve as source input data, reservoir parameters corresponding to the source input data serve as source output data to construct a neural network model, the source input data and the source output data are trained, and a source model is obtained; taking the logging data as target input data, taking the reservoir parameters corresponding to the target input data as target output data, and constructing a neural network model as an original oil and gas prediction model; and migrating the model parameters of the source model to the original oil and gas prediction model, and training the target input data and the target output data to obtain a target oil and gas prediction model. According to the method provided by the invention, the model parameters of the source model are migrated to the original oil and gas prediction model by utilizing migration learning, the infl