Home

Awesome

ONNC (Open Neural Network Compiler)

Introduction

ONNC (Open Neural Network Compiler) is a retargetable compilation framework designed specifically for proprietary deep learning accelerators. Its software architecture expedites porting ONNC to any Deep Learning Accelerator (DLA) design that supports ONNX (Open Neural Network Exchange) operators. ONNC guarantees executability across every DLA by means of transforming ONNX models into DLA-specific binary forms and leveraging the intermediate representation (IR) design of ONNX along with effective algorithms to eliminate the overhead of data movement. ONNC is the first open source compiler available for NVDLA-based hardware designs. Its NVDLA backend can compile a model into an executable NVDLA Loadable file. Integrating ONNC with the NVDLA software stack opens up opportunities for developers and researchers to explore the NVDLA-based inference design at system level.

Papers

Documentation

Current Status

How to contribute

Directory Structure

Supported platforms

ONNC supports Ubuntu/x86_64 and MacOSX.

Here is a list of verified versions:

Getting Started

There are three ways to build ONNC:

  1. Build ONNC via Docker
    Please refer to the ONNC Utilities document.
  2. Build ONNC via ONNC umbrella
    Please follow the instructions of README.md in onnc-umbrella.
    Here is the version of external library we are using in ONNC.
  3. Build ONNC without ONNC umbrella
    Please refer to the ONNC CMake build instruction