摘要
为了满足训练神经网络的计算要求,将浮点数据替换为定点数据,使用更少的比特数,从而提高性能密度。虽然使用定点数据会降低精度,但只要性能比精度更关键,就可以使用定点数据。定点数据类型用于神经网络,可以降低使用FPGA实现神经网络的硬件资源消耗。
In order to meet the computing requirements for training neural networks,floating-point data is replaced with fixed-point data using fewer bits to improve performance density.Although using fixed-point data may reduce accuracy,it can be used as long as performance is more critical than precision.Fixed-point data types are used for neural networks and can reduce the hardware resource consumption of implementing neural networks on FPGA.
作者
陈瑶
王永强
王远飞
邵然
赵思成
Chen Yao;Wang Yongqiang;Wang Yuanfei;Shao Ran;Zhao Sicheng(Harbin Vocational and Technical College,Harbin,China)
出处
《科学技术创新》
2023年第13期78-82,共5页
Scientific and Technological Innovation
基金
哈尔滨职业技术学院校内课题:人工神经元网络硬件实现的研究(HZY2020ZY006)。