importONNXNetwork

Import pretrainedONNXnetwork

Syntax

net = importONNXNetwork(modelfile,'OutputLayerType',outputtype)

net = importONNXNetwork(modelfile,'OutputLayerType',outputtype,'Classes',classes)

Description

net= importONNXNetwork(modelfile,'OutputLayerType',outputtype)imports a pretrained network from the ONNX™ (Open Neural Network Exchange) filemodelfileand specifies the output layer type of the imported network.

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

example

net= importONNXNetwork(modelfile,'OutputLayerType',outputtype,'Classes',classes)additionally specifies the classes for a classification network.

Examples

collapse all

Download and InstallDeep Learning ToolboxConverter forONNX模型格式

This example uses:

Download and install theDeep Learning Toolbox Converter for ONNX Model Formatsupport package.

TypeimportONNXNetworkat the command line.

importONNXNetwork

IfDeep Learning Toolbox Converter for ONNX Model Formatis 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 clickInstall. Check that the installation is successful by importing the network from the model file'cifarResNet.onnx'at the command line. If the support package is installed, then the function returns aDAGNetworkobject.

modelfile ='cifarResNet.onnx'; classes = ["airplane""automobile""bird""cat""dee""dog""frog""horse""ship""truck"]; net = importONNXNetwork(modelfile,'OutputLayerType','classification','Classes',classes)

net = DAGNetwork with properties: Layers: [77×1 nnet.cnn.layer.Layer] Connections: [85×2 table]

Import ONNX Network

This example uses:

Open Live Script

Import a residual neural network trained on the CIFAR-10 data set. Specify the file containing the ONNX network, its output type, and its output classes.

modelfile ='cifarResNet.onnx'; classes = ["airplane""automobile""bird""cat""deer""dog""frog""horse""ship""truck"]; net = importONNXNetwork(modelfile,'OutputLayerType','classification','Classes',classes)

net = DAGNetwork with properties: Layers: [77×1 nnet.cnn.layer.Layer] Connections: [85×2 table] InputNames: {'Input_input'} OutputNames: {'ClassificationLayer_softmax'}

Analyze the imported network.

analyzeNetwork(net)

Input Arguments

collapse all

`modelfile`—的名字ONNX model file
character vector|string scalar

的名字ONNX model file containing the network, specified as a character vector or a string scalar. The file must be in the current folder, in a folder on the MATLAB^®path, or you must include a full or relative path to the file.

Example:'cifarResNet.onnx'

`outputtype`—Type of output layer
`'classification'`|`'regression'`|`'pixelclassification'`

Type of the output layer that the function appends to the end of the imported network, specified as'classification','regression', or'pixelclassification'. Using'pixelclassification'appends apixelClassificationLayer(计算机视觉Toolbox)object (requires Computer Vision Toolbox™).

If a network inmodelfilehas multiple outputs, then you cannot specify the output layer types using this argument. UseimportONNXLayersinstead.importONNXLayersinserts placeholder layers for the outputs. After importing, you can find and replace the placeholder layers by usingfindPlaceholderLayersandreplaceLayer, respectively.

Example:'regression'

`classes`—Classes of the output layer
`'auto'`(default) |categorical vector|string array|cell array of character vectors

Classes of the output layer, specified as a categorical vector, string array, cell array of character vectors, or'auto'. IfClassesis'auto', then the software sets the classes tocategorical(1:N), whereNis the number of classes. If you specify a string array or cell array of character vectorsstr, then the software sets the classes of the output layer tocategorical(str,str).

Data Types:char|categorical|string|cell

Output Arguments

collapse all

`net`— Pretrained network
`DAGNetwork`object

Pretrained network, returned asDAGNetworkobject.

Limitations

importONNXNetworksupports ONNX versions as follows:
- importONNXNetworksupports ONNX intermediate representation version 6.
- importONNXNetworkfully supports ONNX operator sets 6, 7, 8, and 9.
- importONNXNetworkoffers limited support for ONNX operator sets 10 and 11.

Note

If you import an exported network, layers of the reimported network might differ from the original network and might not be supported.

Tips

If the ONNX network contains a layer thatDeep Learning Toolbox Converter for ONNX Model Formatdoes not support, then the function returns an error message. In this case, you can still useimportONNXLayersto import the network architecture and weights.

importONNXNetworksupports the following ONNX layers, with some limitations:

ONNX Layer	Deep Learning Toolbox Layer
`Add`	`additionLayer`或`nnet.onnx.layer.ElementwiseAffineLayer`
`AveragePool`	`averagePooling2dLayer`
`BatchNormalization`	`batchNormalizationLayer`
`Clip`	`nnet.onnx.layer.ClipLayer`
`Concat`	`concatenationLayer`
`Constant`	None (Imported as weights)
`Conv`	`convolution2dLayer`
`ConvTranspose`	`transposedConv2dLayer`
`Div`	`nnet.onnx.layer.ElementwiseAffineLayer`
`Dropout`	`dropoutLayer`
`Flatten`	`nnet.onnx.layer.FlattenLayer`或`nnet.onnx.layer.Flatten3dLayer`
`Elu`	`eluLayer`
`Gemm`	`fullyConnectedLayer`if ONNX network is recurrent, otherwise`nnet.onnx.layer.FlattenLayer`followed by`convolution2dLayer`
`GlobalAveragePool`	`globalAveragePooling2dLayer`
`GlobalMaxPool`	`globalMaxPooling2dLayer`
`GRU`	`gruLayer`
`Identity`	`nnet.onnx.layer.IdentityLayer`
`ImageScaler`	`nnet.onnx.layer.ElementwiseAffineLayer`
`InstanceNormalization`	`groupNormalizationLayer`with`numGroups`specified as`"channel-wise"`
`LeakyRelu`	`leakyReluLayer`
`LRN`	`CrossChannelNormalizationLayer`
`LSTM`	`lstmLayer`或`bilstmLayer`
`MatMul`	`fullyConnectedLayer`if ONNX network is recurrent, otherwise`convolution2dLayer`
`MaxPool`	`maxPooling2dLayer`
`Mul`	`multiplicationLayer`
`PRelu`	`nnet.onnx.layer.PReluLayer`
`Relu`	`reluLayer`或`clippedReluLayer`
`Reshape`	`nnet.onnx.layer.FlattenLayer`
`乙状结肠`	`sigmoidLayer`
`Softmax`	`softmaxLayer`
`Sub`	`nnet.onnx.layer.ElementwiseAffineLayer`
`Sum`	`additionLayer`
`Tanh`	`tanhLayer`

ONNX Layer	Computer Vision Toolbox Layer
`SpaceToDepth`	`spaceToDepthLayer`(计算机视觉Toolbox)

ONNX Layer	Image Processing Toolbox™
`Resize`	`resize2dLayer`(Image Processing Toolbox)或`resize3dLayer`(Image Processing Toolbox)
`Upsample`	`resize2dLayer`(Image Processing Toolbox)或`resize3dLayer`(Image Processing Toolbox)

You can import an ONNX network with multiple inputs and a single output usingimportONNXNetwork. If the network has multiple outputs, useimportONNXLayers. TheimportONNXLayersfunction inserts placeholder layers for the outputs. After importing, you can find and replace the placeholder layers by usingfindPlaceholderLayersandreplaceLayer, respectively. For an example, seeImport ONNX Network with Multiple Outputs. To learn about a deep learning network with multiple inputs and multiple outputs, seeMultiple-Input and Multiple-Output Networks.
To use a pretrained network for prediction or transfer learning on new images, you must preprocess your images in the same way the images that were used to train the imported model were preprocessed. Most common preprocessing steps are resizing images, subtracting image average values, and converting the images from BGR images to RGB.
- To resize images, useimresize. For example,imresize(image,[227,227,3]).
- To convert images from RGB to BGR format, useflip. For example,flip(image,3).
For more information on preprocessing images for training and prediction, seePreprocess Images for Deep Learning.

Compatibility Considerations

expand all

`'ClassNames'`option will be removed

Not recommended starting in R2018b

'ClassNames'will be removed. Use'Classes'instead. To update your code, replace all instances of'ClassNames'with'Classes'. There are some differences between the corresponding properties in classification output layers that require additional updates to your code.

TheClassNamesproperty of a classification output layer is a cell array of character vectors. TheClassesproperty is a categorical array. To use the value ofClasseswith functions that require cell array input, convert the classes using thecellstrfunction.

References

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

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

importONNXNetwork

Syntax

Description

Examples

Download and InstallDeep Learning ToolboxConverter forONNX模型格式

Import ONNX Network

Input Arguments

`modelfile`—的名字ONNX model file
character vector|string scalar

`outputtype`—Type of output layer
`'classification'`|`'regression'`|`'pixelclassification'`

`classes`—Classes of the output layer
`'auto'`(default) |categorical vector|string array|cell array of character vectors

Output Arguments

`net`— Pretrained network
`DAGNetwork`object

Limitations

Tips

Compatibility Considerations

`'ClassNames'`option will be removed

References

See Also

Topics

Deep Learning Toolbox Documentation

万博1manbetx

Introducing Deep Learning with MATLAB

importONNXNetwork

Syntax

Description

Examples

Download and InstallDeep Learning ToolboxConverter forONNX模型格式

Import ONNX Network

Input Arguments

modelfile—的名字ONNX model filecharacter vector|string scalar

outputtype—Type of output layer'classification'|'regression'|'pixelclassification'

classes—Classes of the output layer'auto'(default) |categorical vector|string array|cell array of character vectors

Output Arguments

net— Pretrained networkDAGNetworkobject

Limitations

Tips

Compatibility Considerations

'ClassNames'option will be removed

References

See Also

Topics

Deep Learning Toolbox Documentation

万博1manbetx

Introducing Deep Learning with MATLAB

`modelfile`—的名字ONNX model file
character vector|string scalar

`outputtype`—Type of output layer
`'classification'`|`'regression'`|`'pixelclassification'`

`classes`—Classes of the output layer
`'auto'`(default) |categorical vector|string array|cell array of character vectors

`net`— Pretrained network
`DAGNetwork`object

`'ClassNames'`option will be removed