Model training method, image classification method, server and storage medium

Abstract

The embodiment of the invention provides a model training method, an image classification method, a server and a storage medium. In the embodiment of the invention, the plurality of expert models in the first model comprise at least two expert models with different network structures, so that knowledge learning can be carried out by using different expert models, and samples with different difficulties can be identified when the number of the samples is limited. The fusion model is trained according to the image features extracted from the sample images after the expert models are converged, so that the fusion model can dynamically learn how to reasonably utilize knowledge learned by the multiple expert models according to the knowledge learned by the expert models; therefore, expert models with different performances make different contributions to the finally output classification result, and efficient learning can be carried out even if the number of sample images is small.

Description

technical field [0001] The present application relates to the technical field of computer vision, in particular to a model training, image classification method, server and storage medium. Background technique [0002] The development of neural networks has continuously promoted the progress of artificial intelligence, and the training of neural networks is inseparable from massive training data and pre-trained models. In some application scenarios, it is often faced with the technical problem of lacking sufficient training data to train a sufficiently effective neural network model. Therefore, a solution remains to be proposed. Contents of the invention [0003] Aspects of the embodiments of the present application provide a model training, an image classification method, a server, and a storage medium, which are used to improve the performance of a neural network model in the case of limited training data. [0004] An embodiment of the present application provides an i...

AI Technical Summary

Problems solved by technology

In some application scenarios, it is often faced with the technical problem of lacking sufficient training data to train a sufficiently effective neural network model

Images