Neural network training method, image recognition method, device, equipment and medium

Abstract

The embodiment of the invention discloses a neural network training method, an image recognition method, a device, equipment and a medium. The neural network comprises a plurality of network layers and at least one enhancement layer, the at least one enhancement layer is arranged between the network layers, and the method comprises the following steps: acquiring an original image sample; inputtingthe original image sample into the neural network to obtain a prediction result; acquiring a real result corresponding to the original image sample; determining a loss value according to a loss function, the prediction result and the real result, and training parameters of the neural network based on the loss value; and when a training condition is satisfied, removing the enhancement layer in theneural network to obtain a trained neural network. Not only can the occupancy rate of the samples on the storage space be reduced, but also the generalization performance of the neural network can beimproved.

Description

technical field [0001] Embodiments of the present disclosure relate to the technical field of machine learning, and in particular, to a neural network training method, image recognition method, device, device, and medium. Background technique [0002] Artificial neural networks rely on the training of training samples, that is, labeled image data. At present, data augmentation has become an effective means to solve the sample size. [0003] In related technologies, the means of data augmentation not only occupies a large amount of memory, but also occupies a large amount of input and output (I / O) during the training process, thereby slowing down the training speed. In addition, related technologies directly process images, and some processing methods (such as adversarial processing) cannot be implemented, which cannot effectively improve the diversity of samples, and thus cannot effectively improve the generalization performance of neural networks. Contents of the inventi...

AI Technical Summary

Problems solved by technology

[0003] In related technologies, the means of data expansion not only occupy a large amount of memory, but also occupy a large amount of input and output (I / O) during the training process, thereby slowing down the training speed

In addition, related technologies directly process images, and some processing methods (such as confrontation processing) cannot be implemented, which cannot effectively improve the diversity of samples, and thus cannot effectively improve the generalization performance of neural networks.

Images