Method and device for compressing neural network model

• Main Authors: WANG GUIBIN, JIA LEI, CONG SHIJUN, DONG HAO
• Format: Patent
• Language: Chinese; English
• Published: 29.04.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention provides a method and device for compressing a neural network model, electronic equipment and a readable storage medium, and relates to the technical field of artificial intelligence such as deep learning and cloud service. The method for compressing the neural network model comprises the following steps: acquiring a to-be-compressed neural network model; determining a first bit width, a second bit width and a target sparse rate corresponding to the to-be-compressed neural network model; obtaining a target value according to the first bit width, the second bit width and the target sparse rate; and compressing the to-be-compressed neural network model by using the target value, the first bit width and the second bit width to obtain a compression result of the to-be-compressed neural network model. The compression steps of the neural network model can be simplified, and the compression efficiency of the neural network model can be improved. 本公开提供了一种压缩神经网络模型的方法、装置、电子设备及可读存储介质，涉及深度学习、云服务等人工智能技术领域。其中