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 expressivity, composability, and portability. We present Relay, a new compiler framework for DL. Relays functional, statically typed intermediate representation (IR) unifies and generalizes existing DL IRs to express state-of-the-art models. The introduction of Relays expressive IR requires careful design of domain-specific optimizations, addressed via Relays extension mechanisms. Using these extension mechanisms, Relay supports a unified compiler that can target a variety of hardware platforms. Our evaluation demonstrates Relays competitive performance for a broad class of models and devices (CPUs, GPUs, and emerging accelerators). Relays design demonstrates how a unified IR can provide expressivity, composability, and portability without compromising performance.
Download