首页 | 本学科首页   官方微博 | 高级检索  
     检索      

GPU加速的自适应仿射传播聚类方法
引用本文:陈艳阳,曾卫明.GPU加速的自适应仿射传播聚类方法[J].计算机系统应用,2016,25(11):146-150.
作者姓名:陈艳阳  曾卫明
作者单位:上海海事大学 信息工程学院, 上海 201306,上海海事大学 信息工程学院, 上海 201306
基金项目:国家自然科学基金(31470954)
摘    要:自适应仿射传播聚类作为一种新兴的聚类算法,不需要指定初始类心以及类数,对解决聚类中类数不确定性问题非常有效.然而,自适应仿射传播聚类存在时间消耗过大的问题,当样本数量较大时运行速度缓慢.为了提高自适应仿射传播聚类的运行速度,基于NVIDIA公司的统一计算设备架构(Compute Unified Device Architecture,CUDA)和Matlab并行工具箱,提出了一种自适应仿射传播聚类的并行化方法.实验结果表明,基于GPU并行化的自适应仿射传播聚类在运行速度上有了明显提高,与该算法的串行执行方式相比,运行速度提升2倍以上,并且随着样本数量的增长,加速性能越来越好.

关 键 词:自适应仿射传播聚类  并行化  统一计算设备架构  并行工具箱  GPU加速

GPU-Accelerated Adaptive Affinity Propagation Clustering Method
CHEN Yan-Yang and ZENG Wei-Ming.GPU-Accelerated Adaptive Affinity Propagation Clustering Method[J].Computer Systems& Applications,2016,25(11):146-150.
Authors:CHEN Yan-Yang and ZENG Wei-Ming
Institution:Lab of Digital Image and Intelligent Computation, College of Information Engineering, Shanghai Maritime University, Shanghai 201306, China and Lab of Digital Image and Intelligent Computation, College of Information Engineering, Shanghai Maritime University, Shanghai 201306, China
Abstract:Adaptive affinity propagation clustering(adaptive Affinity propagation clustering, adAP), as a new clustering algorithm, does not need to specify the initial "exemplars" and the class number, which is effective to solve the problem of class number uncertainty in clustering. Then, as a result of the adAP is extremely time consuming, the larger the number of samples is, the slower the speed is. In order to improve the speed of the adAP, this paper realizes a parallel method, which is based on NVIDIA''s Compute Unified Device Architecture (CUDA) and Matlab parallel computing toolbox. The experiment results show that the GPU-based parallel adAP method has a certain speedup effect, and it is more than 2 times faster than the serial execution. With the increase of the number of samples, the acceleration performance is getting better and better.
Keywords:adaptive affinity propagation clustering  parallelization  compute unified device architecture  parallel computing toolbox  GPU acceleration
点击此处可从《计算机系统应用》浏览原始摘要信息
点击此处可从《计算机系统应用》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号