id	author	title	date	pages	extension	mime	words	sentence	flesch	summary	cache	txt
ajst-20214	Wang, Zeyang; Mei, Huawei; Fang, Yi	A Design of Model Compression Method Based on YOLOv5	2024	6	.pdf	application/pdf	3940	229	50	Therefore, many experts and scholars have begun to investigate methods for lightweighting deep learning models and deploying lightweight target detection models on edge devices to enable real-time detection tasks. Specifically, BN layers normalize the distribution of input data by standardizing each feature dimension to have a mean of 0 and a variance of 1, thus avoiding gradient vanishing issues caused by deeper network layers and speeding up the convergence of model training.	cache/ajst-20214.pdf	txt/ajst-20214.txt