首页> 外国专利> SPARSE MATRIX NORMALIZATION DEVICE, SPARSE MATRIX NORMALIZATION METHOD, SPARSE MATRIX NORMALIZATION PROGRAM, AND DATA STRUCTURE

SPARSE MATRIX NORMALIZATION DEVICE, SPARSE MATRIX NORMALIZATION METHOD, SPARSE MATRIX NORMALIZATION PROGRAM, AND DATA STRUCTURE

机译:稀疏矩阵归一化设备,稀疏矩阵归一化方法,稀疏矩阵归一化程序和数据结构

摘要

A sparse matrix normalization device 10 is provided with: a calculation unit 11 which performs, for each column of a target matrix to be normalized, a calculation process to calculate the average and standard deviation of the value of each component of the column; a first dividing unit 12 which performs, for each column of the target matrix, a first dividing process to divide the value of each component of the column by the standard deviation calculated on the basis of the column; a second dividing unit 13 which performs, for each column of the target matrix, a second dividing process to divide the average calculated on the basis of the column by the standard deviation calculated on the basis of the column; and a generation unit 14 which arranges the quotients calculated by a plurality of second dividing processes, in a row in the order of the columns of the target matrix from which the quotients are calculated, thereby generating a row vector.
机译:稀疏矩阵归一化装置10包括:计算单元11,其对要归一化的目标矩阵的每一列执行计算过程,以计算该列的每个分量的值的平均值和标准偏差。第一除法单元12,对目标矩阵的每一列执行第一除法处理,以将该列的每个分量的值除以基于该列计算的标准差;第二除法单元13对目标矩阵的每一列执行第二除法处理,以将基于该列计算的平均值除以基于该列计算的标准偏差;生成单元14,将通过多个第二除法处理而得到的商,按照计算出商的目标矩阵的列的顺序排列在一行中,从而生成行向量。

著录项

  • 公开/公告号WO2020115844A1

    专利类型

  • 公开/公告日2020-06-11

    原文格式PDF

  • 申请/专利权人 NEC CORPORATION;

    申请/专利号WO2018JP44765

  • 发明设计人 ARAKI TAKUYA;

    申请日2018-12-05

  • 分类号G06F17/16;

  • 国家 WO

  • 入库时间 2022-08-21 11:10:47

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号