Linux Foundation's AI Foundation (LF AI) is working to build an ecosystem to maintain artificial intelligence (AI), machine learning (ML) and deep learning (DL) open source innovation. The organization recently announced , open neural network exchange (ONNX) has become the latest of these projects.
ONNX originally developed by Microsoft and Facebook in 2017 and open source, designed to make it easier to create and deploy machine learning applications. ONNX provide depth learning model and the traditional machine learning models generic forms, in order to achieve a framework, compiler, interoperability between runtime and hardware accelerators. ONNX also has a number of advantages, including portability, it makes the AI developers can more easily move between AI model trusted AI / ML / DL workflow tool.
ONNX format has become a norm and standards, it is not only company-approved norms, and has been actively implemented in the product. Including Amazon, Facebook, Intel, Microsoft, NVIDIA and other companies are ONNX active contributor to the code base.
A year ago, Microsoft's open source ONNX runs, it is an inference engine for high-performance machine learning models ONNX format. But the addition of LF AI Foundation ONNX format only, does not include ONNX running.
"We are delighted to welcome ONNX join LF AI Foundation. We will ONNX as a key project of open source AI continued growth," Chairman of the LF AI Foundation Council Mazin Gilbert representation. "We are committed to expanding the open-source AI, trust and support, transparency and accessibility of community development."
ERIC BOYD vice president of Microsoft's Azure AI company in the announcement wrote, "We are proud of the progress made ONNX and want recognition of the contribution, ideas and enthusiasm throughout ONNX community. We are excited about the future ONNX and all."