ONNX-Go, neural networks made easy
summary
A Go package that reads and executes ONNX models and exposes them as regular Go values, relying on an execution backend (cornea is used as the working backend). Workflow is concrete: create a backend, load a binary ONNX model, bind inputs, call compute and read outputs. Current status is explicit: proof of concept works end-to-end — author imports a pre-trained MNIST model, feeds an image and gets a prediction — and needs contributors to stabilize and add backends.
AI-written summary · about
related talks
05mONNX Runtime
45mLeverage Power of Machine Learning with ONNX
41mLeverage Power of Machine Learning with ONNX
51mMaking neural networks run in browser with ONNX
45mMaking Machine Learning Applications Fast and Simple with...
30mScalable ML acceleration with ONNX Runtime
30mThe Evolution of ONNX: Driving Interoperability in AI...
33mWhat on earth is ONNX and ONNXRUNTIME