Method and device for obtaining recognition model, electronic equipment and storage medium

Abstract

The invention relates to a method and device for obtaining an identification model, electronic equipment and a storage medium, and belongs to the technical field of deep learning, and the method comprises the steps: determining a first identity category to which each piece of sample data belongs according to a first identification model in the process of training the identification model; according to a first identity category and a second identity category to which the sample data belongs, determining the weight of a first probability in the model training process, the first probability beingthe probability that the sample data belongs to the second identity category, and the second identity category being the identity category to which the sample data actually belongs; and performing iterative training on the first recognition model according to the first probability and the weight of the sample data to obtain a second recognition model. According to the method and the device, the loss value of the sample data is calculated through the weighted first probability to carry out model training, and the correlation between classes in the model training process is fully considered, sothat the accuracy of identifying the model is improved.

Description

technical field [0001] The present disclosure relates to the technical field of deep learning, and in particular to a method, device, electronic device and storage medium for acquiring a recognition model. Background technique [0002] With the development of artificial intelligence, the application of deep learning technology is becoming more and more extensive. Especially when solving the problem of identifying identity information from images, the image is usually identified according to the image recognition model. Therefore, model training is usually required to establish an image recognition model before image recognition. For example, before using an image recognition model to recognize the age corresponding to a face in a face image, model training is required to establish a face age recognition model. [0003] In related technologies, in the process of training the image recognition model, the age is divided into several age groups, and the probability of multiple...

AI Technical Summary

Problems solved by technology

[0005] The present disclosure provides a method, device, electronic device, and storage medium for obtaining a recognition model, which can overcome the problem of model training directly based on the probability of the plurality of sample images in each age group when training the recognition model, which may cause model The training is not accurate, which leads to the problem of low accuracy of the trained recognition model

Images