exportONNXNetwork

Export network to ONNX model format

Syntax

exportONNXNetwork(net,filename)

exportONNXNetwork(net,filename,Name,Value)

Description

Export a trained Deep Learning Toolbox™ network to the ONNX™ (Open Neural Network Exchange) model format.

exportONNXNetwork(net,filename) exports the deep learning network net with weights to the ONNX format file filename. If filename exists, then exportONNXNetwork overwrites the file.

This function requires the Deep Learning Toolbox Converter for ONNX Model Format support package. If this support package is not installed, then the function provides a download link.

exportONNXNetwork(net,filename,Name,Value) exports a network using additional options specified by one or more name-value pair arguments.

Examples

Export Network in ONNX Format

Load a pretrained SqueezeNet convolutional neural network. If Deep Learning Toolbox Model for SqueezeNet Network is not installed, then the squeezenet function provides a download link.

net = squeezenet

net = 
  DAGNetwork with properties:

         Layers: [68×1 nnet.cnn.layer.Layer]
    Connections: [75×2 table]

Export the network as an ONNX format file in the current folder called squeezenet.onnx. If the Deep Learning Toolbox Converter for ONNX Model Format support package is not installed, then the function provides a link to the required support package in the Add-On Explorer. To install the support package, click the link, and then click Install.

filename = 'squeezenet.onnx';
exportONNXNetwork(net,filename)

Now, you can import the squeezenet.onnx file into any deep learning framework that supports ONNX import.

Input Arguments

net — Trained network

SeriesNetwork object | DAGNetwork object

Trained network, specified as a SeriesNetwork or a DAGNetwork object. You can get a trained network by importing a pretrained network (for example, by using the googlenet function) or by training your own network using trainNetwork.

filename — Name of file

character vector | string scalar

Name of file, specified as a character vector or string scalar.

Example: 'network.onnx'

Name-Value Pair Arguments

Specify optional comma-separated pairs of Name,Value arguments. Name is the argument name and Value is the corresponding value. Name must appear inside quotes. You can specify several name and value pair arguments in any order as Name1,Value1,...,NameN,ValueN.

Example: exportONNXNetwork(net,filename,'NetworkName','my_net') exports a network and specifies 'my_net' as the network name in the saved ONNX network.

'NetworkName' — Name of ONNX network

'Network' (default) | character vector | string scalar

Name of ONNX network to store in the saved file, specified as a character vector or a string scalar.

Example: 'my_squeezenet'

'OpsetVersion' — Version of ONNX operator set

8 (default) | 6 | 7 | 9

Version of ONNX operator set to use in the exported model. If the default operator set does not support the network you are trying to export, then try using a later version. If you import the exported network to another framework and you used an operator set during export that the importer does not support, then the import can fail.

Example: 6

Tips

• exportONNXNetwork does not export settings or properties related to network training such as training options, learning rate factors, or regularization factors.

• If you export a network that contains a layer that the ONNX format does not support, then exportONNXNetwork saves a placeholder ONNX operator in place of the unsupported layer and returns a warning. You cannot import an ONNX network with a placeholder operator into other deep learning frameworks.

• exportONNXNetwork can export the following:

  • Networks that have both convolutional and LSTM layers, for example, for video classification applications.

  • All custom layers that are created when importing networks from ONNX or TensorFlow™-Keras using Deep Learning Toolbox Converter for ONNX Model Format or Deep Learning Toolbox Importer for TensorFlow-Keras Models as in the below table.

  • The following layers:

    ONNX Exporter Supported Layers
    Deep Learning Toolbox Layers
    additionLayer
    averagePooling2dLayer
    averagePooling3dLayer
    batchNormalizationLayer
    bilstmLayer
    classificationLayer
    clippedReluLayer
    convolution2dLayer
    concatenationLayer
    convolution3dLayer
    crop2dLayer
    CrossChannelNormalizationLayer
    depthConcatenationLayer
    dropoutLayer
    fullyConnectedLayer
    flattenLayer
    globalAveragePooling2dLayer
    groupedConvolution2dLayer
    imageInputLayer
    image3dInputLayer
    leakyReluLayer
    lstmLayer
    maxPooling2dLayer
    maxPooling3dLayer
    maxUnpooling2dLayer
    regressionLayer
    reluLayer
    sequenceInputLayer
    softmaxLayer
    transposedConv2dLayer
    transposedConv3dLayer
    ONNX Importer Custom Layers
    nnet.onnx.layer.ClipLayer
    nnet.onnx.layer.FlattenLayer
    nnet.onnx.layer.GlobalAveragePooling2dLayer
    nnet.onnx.layer.IdentityLayer
    nnet.onnx.layer.PReluLayer
    nnet.onnx.layer.ElementwiseAffineLayer
    nnet.onnx.layer.TanhLayer
    Keras Importer Custom Layers
    nnet.keras.layer.FlattenCStyleLayer
    nnet.keras.layer.GlobalAveragePooling2dLayer
    nnet.keras.layer.SigmoidLayer
    nnet.keras.layer.TanhLayer
    nnet.keras.layer.ZeroPadding2dLayer
    Caffe Importer Custom Layers
    nnet.caffe.layer.TanhLayer
    Computer Vision Toolbox™ Layers
    pixelClassificationLayer
    rcnnBoxRegressionLayer
    regionProposalLayer
    roiInputLayer
    roiMaxPooling2dLayer
    rpnClassificationLayer
    rpnSoftmaxLayer
    Text Analytics Toolbox™ Layers
    wordEmbeddingLayer

References

[1] Open Neural Network Exchange. https://github.com/onnx/.

[2] ONNX. https://onnx.ai/.