Microsoft contributes the open neural network exchange to the Linux Foundation
The AI Foundation (LF AI), part of the Linux Foundation, is working to create an ecosystem that sustains open source innovation in artificial intelligence (AI), machine learning (ML), and deep learning (DL). Recently, the organization announced that Open Neural Network Exchange (ONNX) has become the latest project.
Originally developed and open-sourced by Microsoft and Facebook in 2017, ONNX is designed to make it easier to create and deploy machine learning applications. ONNX provides a common representation of deep learning models and traditional machine learning models, enabling interoperability between frameworks, compilers, runtimes, and hardware accelerators. ONNX also has many advantages, including portability, which makes it easier for AI developers to move AI models between tools in trusted AI/ML/DL workflows.
The ONNX has gradually become a specification and standard. It is not only a company-approved specification but has also been actively implemented in the product. Companies such as Amazon, Facebook, Intel, Microsoft, and NVIDIA are all active contributors to the ONNX code base.
ERIC BOYD, vice president of Microsoft Azure AI, wrote in the announcement:
We are excited that LF AI will host ONNX and continue the open-governance model, which encourages community participation and contributions. LF AI will provide long term leadership to ONNX enabling a community focused on accelerating the adoption of machine learning and fostering ONNX’s next wave of innovation and adoption.
