基于多线程GPU的CNN图像分类实现

2017 International SoC Design Conference (ISOCC) Pub Date : 2017-11-01 DOI:10.1109/ISOCC.2017.8368904

Seong-Hyeon Han, Kwang-Yeob Lee

{"title":"基于多线程GPU的CNN图像分类实现","authors":"Seong-Hyeon Han, Kwang-Yeob Lee","doi":"10.1109/ISOCC.2017.8368904","DOIUrl":null,"url":null,"abstract":"This study implemented an image classification CNN using a multi-thread GPU. For the CNN, the CIFAR10 dataset was used, and the multi-thread GPU had 256 threads. Using the 256 threads limited to each layer, allocation and parallel processing were conducted. The image classification CNN took 807 ms for computation.","PeriodicalId":248826,"journal":{"name":"2017 International SoC Design Conference (ISOCC)","volume":"500 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":"{\"title\":\"Implemetation of image classification CNN using multi thread GPU\",\"authors\":\"Seong-Hyeon Han, Kwang-Yeob Lee\",\"doi\":\"10.1109/ISOCC.2017.8368904\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This study implemented an image classification CNN using a multi-thread GPU. For the CNN, the CIFAR10 dataset was used, and the multi-thread GPU had 256 threads. Using the 256 threads limited to each layer, allocation and parallel processing were conducted. The image classification CNN took 807 ms for computation.\",\"PeriodicalId\":248826,\"journal\":{\"name\":\"2017 International SoC Design Conference (ISOCC)\",\"volume\":\"500 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"9\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 International SoC Design Conference (ISOCC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISOCC.2017.8368904\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 International SoC Design Conference (ISOCC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISOCC.2017.8368904","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 9

摘要

本研究使用多线程GPU实现了一种图像分类CNN。对于CNN，使用CIFAR10数据集，多线程GPU有256个线程。利用每层限定的256个线程，进行分配和并行处理。图像分类CNN的计算时间为807 ms。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Implemetation of image classification CNN using multi thread GPU

This study implemented an image classification CNN using a multi-thread GPU. For the CNN, the CIFAR10 dataset was used, and the multi-thread GPU had 256 threads. Using the 256 threads limited to each layer, allocation and parallel processing were conducted. The image classification CNN took 807 ms for computation.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2017 International SoC Design Conference (ISOCC)

自引率

0.00%

发文量