高技术通讯2023，Vol.33Issue(11)：1160-1171,12.DOI:10.3772/j.issn.1002-0470.2023.11.004
基于自适应静态数据布局策略的深度学习张量程序自动生成框架
A deep learning tensor program automatic generation framework based on adaptive layout of static data
摘要
Abstract
How to determine the layout of static/const data is a big challenge faced by tensor program automatic genera-tion frameworks.Ansor,the most broadly-used and promising framework among them,solves this issue by training a performance cost model according to a layout strategy specified in advance,then searching the tensor program with the optimal performance based on the cost model.However,there are two problems:a single strategy cannot be suitable for all tasks,and the performance cost model is not accurate.In order to solve these problems,AL-An-sor,a tensor program automatic generation framework based on the adaptive layout(AL)strategy of static data,is proposed.It adaptively chooses multiple layout strategies during the search process,and trains the performance cost model according to them.In this way,AL-Ansor can find a tensor program with higher performance.Taking convo-lutional layers as workloads,this work evaluates Ansor and AL-Ansor in a target server with a 32-core Intel Xeon CPU.The experimental results show that AL-Ansor improves the execution performance by 13.81%,12.41%,and 16.59%,respectively,on average,compared against Ansor with three specified layout strategies.关键词
深度学习/张量程序自动生成框架/静态数据布局策略/自适应策略/性能预测模型Key words
deep learning/tensor program automatic generation framework/layout of static/const data/adap-tive strategy/performance cost model引用本文复制引用
樊哲,南子渊,郝一帆,杜子东,陈云霁..基于自适应静态数据布局策略的深度学习张量程序自动生成框架[J].高技术通讯,2023,33(11):1160-1171,12.基金项目
国家重点研发计划(2020AAA0103802),国家自然科学基金(61925208,61732020,U19B2019),中国科学院战略性先导科技专项(XDB32050200),北京智源人工智能研究院以及北京市科技新星计划(Z191100001119093),中国科学院稳定支持基础研究领域青年团队计划(YSBR-029)和中国科学院青年创新促进会和科学探索奖资助项目. （2020AAA0103802）
