Feature extraction method and device and computer readable storage medium

• Main Authors: WANG JIANMING, HUANG BO, WU ZHENYU, BI YE
• Format: Patent
• Language: Chinese; English
• Published: 10.09.2019
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a feature extraction method, and the method comprises the steps of obtaining the training data which comprises an original feature corresponding to each piece of sample data; training an initial feature extraction model by using the training data, and obtaining the parameter values of the initial feature extraction model; screening the parameter values of the initial feature extraction model to obtain the screened parameter values; reconstructing the initial feature extraction model by using the screened parameter values to obtain a reconstructed feature extraction model; inputting the training data into the reconstructed feature extraction model to obtain the derived features of each piece of sample data; and re-training the reconstructed feature extraction model according to the derivative features of each piece of sample data and the original features corresponding to each piece of sample data until iteration is ended, and obtaining a trained feature extraction model. The method can