Event Details

Understanding TensorIR: An Abstraction for Tensorized Program Optimization

Date: 12/15/2021 11:00 am
User Tutorial

Organization: Shanghai Jiao Tong University
Speakers: Siyuan Feng
Download Slides

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 tensor computations. These new acceleration primitives, along with the emerging machine learning models, bring tremendous engineering challenges. In this talk, I will introduce TensorIR, a TVM 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.

Register for TVMCon 2021