【微课堂】Accelerate R applications by cuda on GPU
分享题目:Accelerate R applications by cuda on GPU
分享内容:
本次讲座会介绍在数据分析中用到的并行化技术,并以R为平台介绍如何开发移植和优化数据分析的各种算法,如深度学习,knn算法。GPU当今已成为各大主流数据分析平台的必备元素,如Google的tenserflow,aliyun的数据分析云,在本次讲座中,我们还将讲解如何快速利用gpu加速现有的算法,以及各种使用技巧。
分享嘉宾:
赵鹏先生
I am a Senior GPU Architect in HPC and Big Data (Machine Learning) field at NVIDIA. I have eight years of experience developing scientific and engineering applications and is experienced in multi-processes and multi-threading parallelization, algorithm optimization and hardware tuning. Currently, I am working on Modular Dynamic and Machine Learning projects, e.g. cuDNN. Before joining NVIDIA, I worked on the distributed processing and algorithm optimization for EDA software at Cadence.
Specialties: high performance computing, high level algorithmic optimization and low level hardware-specific hot spots tuning, simd, vectorization, code parallelization, computational fluid dynamics(CFD), lattice Boltzmann method(LBM), finite difference method(FDM), partial differential equations(PDE).
Currently, I am running a website for parallelization of R by NVIDIA GPU as below : www.parallelr.com
时间和地点:2016年4月21日(星期四) 晚上9点到10点半 数据人网微课堂。
如何加入微课堂:添加小编微信,luqin360,注明:星期四听课。
伙伴们,数据人网(http://shujuren.org)第二个版本已经上线了,诚邀大家去体验、投稿和做数据知识分享。数据人网也推出一些有利于数据人的活动,介绍如下:
【工作帮】,专注于数据人的职业服务
【微课堂】,专注于数据大咖的知识分享
知道数据人更多有趣的事情,请点击阅读原文。
请关注“恒诺新知”微信公众号,感谢“R语言“,”数据那些事儿“,”老俊俊的生信笔记“,”冷🈚️思“,“珞珈R”,“生信星球”的支持!