版权说明 操作指南
首页 > 成果 > 详情

A pipeline computing method of SpTV for three-order tensors on CPU and GPU

认领
导出
Link by DOI
反馈
分享
QQ微信 微博
成果类型:
期刊论文
作者:
Yang, Wangdong*;Li, Kenli;Li, Keqin
通讯作者:
Yang, Wangdong
作者机构:
[Li, Keqin; Yang, Wangdong; Li, Kenli] Hunan Univ, Coll Comp Sci & Elect Engn, Lushan Rd S, Changsha, Peoples R China.
[Yang, Wangdong] Hunan City Univ, Coll Informat & Elect Engn, 518 Yingbin East Rd, Yiyang, Peoples R China.
[Li, Keqin] SUNY Albany, 353 Broadway, Albany, NY 12246 USA.
通讯机构:
[Yang, Wangdong] H
Hunan Univ, Coll Comp Sci & Elect Engn, Lushan Rd S, Changsha, Peoples R China.
Hunan City Univ, Coll Informat & Elect Engn, 518 Yingbin East Rd, Yiyang, Peoples R China.
语种:
英文
关键词:
Pipeline;Slices;Tensor;Tensor and vector multiplication
期刊:
ACM Transactions on Knowledge Discovery from Data
ISSN:
1556-4681
年:
2019
卷:
13
期:
6
页码:
63:1-63:27
基金类别:
National Key R&D Program of China [2018YFB1003401]; National Natural Science Foundation of ChinaNational Natural Science Foundation of China [61872127, 61572175, 61751204, 61472124]; National Outstanding Youth Science Program of National Natural Science Foundation of ChinaNational Natural Science Foundation of China [61625202]; Key Program of National Natural Science Foundation of ChinaNational Natural Science Foundation of China [61432005]; Natural Science Foundation of Hunan Province, ChinaNatural Science Foundation of Hunan Province [2018JJ2022]
机构署名:
本校为通讯机构
院系归属:
信息与电子工程学院
摘要:
Tensors have drawn a growing attention in many applications, such as physics, engineering science, social networks, recommended systems. Tensor decomposition is the key to explore the inherent intrinsic data relationship of tensor. There are many sparse tensor and vector multiplications (SpTV) in tensor decomposition. We analyze a variety of storage formats of sparse tensors and develop a piecewise compression strategy to improve the storage efficiency of large sparse tensors. This compression strategy can avoid storing a large number of empty ...

反馈

验证码:
看不清楚,换一个
确定
取消

成果认领

标题:
用户 作者 通讯作者
请选择
请选择
确定
取消

提示

该栏目需要登录且有访问权限才可以访问

如果您有访问权限,请直接 登录访问

如果您没有访问权限,请联系管理员申请开通

管理员联系邮箱:yun@hnwdkj.com