Image classification neural network architecture search method based on social evolution algorithm

The invention discloses a neural network architecture search method based on a social evolution algorithm, relates to the field of image classification neural network architecture search, and aims to improve the design efficiency of an image classification network architecture and facilitate discovery of a more complex and efficient neural network architecture. By analyzing the experience of human social development, the individual quality of a continuously developed and optimized society should be better and better, and a social evolution algorithm is proposed for solving the problem of image classification neural network architecture design by referring to the social development. An individual screening operation, a group organization attack operation, an individual innovation operation, a failure tolerance mechanism and a development resource supply operation in the image classification neural network are designed. According to the method, the image neural network does not need to be trained, so that the s