Relay: A High-Level IR for Deep Learning

04/17/2019
by   Jared Roesch, et al.
0

Frameworks for writing, compiling, and optimizing deep learning (DL) models have recently enabled progress in areas like computer vision and natural language processing. Extending these frameworks to accommodate the rapidly diversifying landscape of DL models and hardware platforms presents challenging tradeoffs between expressiveness, composability, and portability. We present Relay, a new intermediate representation (IR) and compiler framework for DL models. The functional, statically-typed Relay IR unifies and generalizes existing DL IRs and can express state-of-the-art models. Relay's expressive IR required careful design of the type system, automatic differentiation, and optimizations. Relay's extensible compiler can eliminate abstraction overhead and target new hardware platforms. The design insights from Relay can be applied to existing frameworks to develop IRs that support extension without compromising on expressivity, composibility, and portability. Our evaluation demonstrates that the Relay prototype can already provide competitive performance for a broad class of models running on CPUs, GPUs, and FPGAs.

READ FULL TEXT
research
04/17/2019

Relay: A High-Level Compiler for Deep Learning

Frameworks for writing, compiling, and optimizing deep learning (DL) mod...
research
09/26/2018

Relay: A New IR for Machine Learning Frameworks

Machine learning powers diverse services in industry including search, t...
research
08/03/2022

HirFuzz: Detecting High-Level Optimization Bugs in DL Compilers via Computational Graph Generation

Deep Learning (DL) compilers are widely adopted to optimize advanced DL ...
research
12/01/2019

PointEval: On the Impact of Pointer Analysis Frameworks

Pointer analysis is a foundational analysis leveraged by various static ...
research
12/10/2019

RVSDG: An Intermediate Representation for Optimizing Compilers

Intermediate Representations (IRs) are central to optimizing compilers a...
research
12/12/2022

Tydi-lang: a language for typed streaming hardware – A manual for future Tydi-lang compiler developers

Transferring composite data structures with variable-length fields often...
research
06/10/2020

Retrofitting Symbolic Holes to LLVM IR

Symbolic holes are one of the fundamental building blocks of solver-aide...

Please sign up or login with your details

Forgot password? Click here to reset