exportONNXNetwork — Export supported feed-forward deep-learning networks to ONNX.
exportONNXNetwork(net, filename) writes deterministic ONNX graph files for RunMat dlnetwork, SeriesNetwork, and DAGNetwork compatibility objects that use the supported sequential feed-forward execution subset.
Syntax
exportONNXNetwork(net, filename, Name, Value)How exportONNXNetwork works
- Supported layers are
featureInputLayer,fullyConnectedLayer,reluLayer,eluLayer,softmaxLayer, and terminalclassificationLayerorregressionLayerobjects. fullyConnectedLayerexports as ONNXGemm; ReLU, ELU, and Softmax export as their corresponding ONNX operators.- Learned parameters are serialized as ONNX DOUBLE tensor initializers using deterministic names derived from layer names.
OpsetVersionsupports versions 13 through 20;BatchSize,InputNames,OutputNames, andVerboseare accepted for single-input/single-output exports.- Unsupported layers, options, or malformed network metadata raise deterministic errors without writing partial graphs.
GPU memory and residency
ONNX export is host graph serialization and file I/O. GPU-resident top-level values are gathered before export; no provider kernels are applicable.
Example
Export ONNX
exportONNXNetwork(net, 'model.onnx', 'OpsetVersion', 14)Expected output:
model.onnx is written.Using exportONNXNetwork with coding agents
Open a RunMat example with live inputs, then ask the agent to explain how exportONNXNetwork changes the result.
Run a small exportONNXNetwork example, explain the result, then change one input and compare the output.
Related Deep Learning functions
adamupdate · analyzeNetwork · bilstmLayer · classificationLayer · combvec · convolution1dLayer · crossentropy · dlarray · dlfeval · dlgradient · dlnetwork · dlupdate · eluLayer · featureInputLayer · forward · fullyConnectedLayer · globalAveragePooling1dLayer · layerGraph · layerNormalizationLayer · lstmLayer · padsequences · regressionLayer · reluLayer · sequenceInputLayer · softmaxLayer · trainingOptions · trainnet · trainNetwork
About RunMat
RunMat is an open-source runtime that executes MATLAB-syntax code blazing on any GPU. It is licensed under the Apache 2.0 license.
- RunMat automatically optimizes your math for GPU execution on Apple, Nvidia, and AMD hardware. No code changes needed. Simulations that took hours now take minutes.
- Start running code in seconds. RunMat runs in the browser, on the desktop, or from the CLI. No license server, no IT ticket.