TensorIR: An Abstraction for Automatic Tensorized Program Optimization | Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2
Authors:
Top Cited Papers
Open Access
- 27 January 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 804-817
Abstract
Deploying deep learning models on various devices has become an important topic. The wave of hardware specialization brings a diverse set of acceleration primitives for multi-dimensional ten- sor computations. These new acceleration primitives, along with the emerging machine learning models, bring tremendous engineering challenges. In this paper, we present TensorIR, a compiler abstraction for optimizing programs with these tensor computation primitives. TensorIR generalizes the loop nest representation used in existing machine learning compilers to bring tensor computation as the first-class citizen. Finally, we build an end-to-end framework on top of our abstraction to automatically optimize deep learning models for given tensor computation primitives. Experimental results show that TensorIR compilation automatically uses the tensor computation primitives for given hardware backends and delivers performance that is competitive to state-of-art hand-optimized systems across platforms.Keywords
This publication has 19 references indexed in Scilit:
- Triton: an intermediate language and compiler for tiled neural network computations2019
- Tiramisu: A Polyhedral Compiler for Expressing Fast and Portable Code2019
- The tensor algebra compilerProceedings of the ACM on Programming Languages, 2017
- In-Datacenter Performance Analysis of a Tensor Processing Unit2017
- XGBoost2016
- Deep Residual Learning for Image Recognition2016
- When polyhedral transformations meet SIMD code generation2013
- HalideACM SIGPLAN Notices, 2013
- Polyhedral Code Generation in the Real WorldLecture Notes in Computer Science, 2006
- The Hierarchically Tiled Arrays programming approach2004