SubscribeFaith Xu, a Senior PM in the Microsoft ML Platform team, brings us up to speed on the Open Neural Network eXchange (ONNX) specification and it's associated Runtime which can be used for running interoperable ML models in Azure.
Media file: https://azpodcast.blob.core.windows.net/episodes/Episode293.mp3
Transcript: https://eus2.videoindexer.ai/accounts/e0eee289-7730-4999-978b-eb7f63be8cb5/videos/77672a62af/?location=EUS2
Resources:
ONNX: https://github.com/onnx/onnx
ONNX Runtime: https://github.com/microsoft/onnxruntime
ONNX Tutorials: https://github.com/onnx/tutorials
ONNX Runtime Tutorials: https://github.com/microsoft/onnxruntime#examples-and-tutorials
ONNX Converters: https://github.com/onnx/onnxmltools/tree/master/onnxmltools
ONNX Ecosystem Docker Image: https://github.com/onnx/onnx-docker/tree/master/onnx-ecosystem
ONNX Runtime in AzureML: https://aka.ms/onnxnotebooks
Other updates:
https://azure.microsoft.com/en-us/updates/public-preview-of-managed-identity-for-azure-service-fabric-applications/
