Interview Speaker
summary
Demonstrates that Microsoft provides a C# machine learning stack ML.NET v1.4 (about 1.5 years old) focused on classic algorithms and now edging into neural nets, explains the common workflow of training in Python/TensorFlow then exporting to ONNX export to run fast in production, and notes hardware vendors like NVIDIA consume ONNX and translate models to GPU shaders so you get real inference speed for C# production inference rather than slow Python-only deployment.
AI-written summary · about
related talks
47mArchitecting for ML.NET & Cognitive Services
41mLeverage Power of Machine Learning with ONNX
07mPyTorch at Microsoft
44mAI for every developer
50mA Pragmatic Deep Dive into ML.NET
58mDeep Learning in Microsoft Azure: CNTK, CaffeOnSpark and Tensorflow
01h 04mIntroduction to ML.NET
45mLeverage Power of Machine Learning with ONNX