Evaluation of sports training effect based on GABP neural network and artificial intelligence

Abstract

Sports training effect evaluation has a direct role in improving the effect of sports training. Based on the idea of artificial intelligence and the GABP neural network algorithm, this paper constructs a sports training effect evaluation model based on the GABP neural network. Moreover, this article vectorizes the code based on the sample matrix to improve code conciseness and increase computational efficiency. In addition, this article tests and evaluates the two neural networks that have been optimized for training, and analyzes and compares the nature of the network, topology, and parameter update methods. After testing, it is found that the trained GABP neural network can achieve the ideal high fitting accuracy of training samples and high generalization ability of test samples under the conditions of low memory usage and short training time. The experimental research results show that the model constructed in this paper meets the actual needs of sports training.