It is right that the extraction of features from data in deep learning is automatic. Training deep learning model is a satisfactory model which carries the training model for various different scenarios and can help to get a better access to business value. The biggest difference between deep learning and traditional pattern recognition methods is that it learns features automatically from big data rather than using hand-designed features. Good features can greatly improve the performance of pattern recognition systems. In various applications of pattern recognition over the past decades, hand-designed features have been dominant. It mainly relies on the designer's a priori knowledge and it is difficult to take advantage of big data. Due to the reliance on hand-tuned parameters, only a small number of parameters are allowed in the design of features.