The system includes input data generation, feature extraction network, credit rating network and sampling rate output network. The model is used to evaluate the credit rating and calculate the sampling rate automatically, which reduces the responsibility of customs clearance and improves the efficiency. At that same time, the intelligent grade and sampling rate compute system of the big data of the cargo customs clearance is used to compute the indexes of the cargo customs clearance of the enterprise custom in recent years, Through the analysis and collation of big data, the credit rating andsampling rate calculation results are more accurate and reasonable, ensuring different goods, different enterprises targeted sampling inspection.