Networks and Layers Supported for C++ Code Generation

MATLAB^® Coder™ supports code generation for series and directed acyclic graph (DAG) convolutional neural networks (CNNs or ConvNets). You can generate code for any trained convolutional neural network whose layers are supported for code generation. See Supported Layers.

Supported Pretrained Networks

The following pretrained networks, available in Deep Learning Toolbox™, are supported for code generation.

Network Name	Description	ARM^® Compute Library	Intel^® MKL-DNN
`AlexNet`	AlexNet convolutional neural network. For the pretrained AlexNet model, see `alexnet`.	Yes	Yes
`DarkNet`	DarkNet-19 and DarkNet-53 convolutional neural networks. For the pretrained DarkNet models, see `darknet19` and `darknet53`.	Yes	Yes
`DenseNet-201`	DenseNet-201 convolutional neural network. For the pretrained DenseNet-201 model, see `densenet201`.	Yes	Yes
`GoogLeNet`	GoogLeNet convolutional neural network. For the pretrained GoogLeNet model, see `googlenet`.	Yes	Yes
`Inception-ResNet-v2`	Inception-ResNet-v2 convolutional neural network. For the pretrained Inception-ResNet-v2 model, see `inceptionresnetv2`.	Yes	Yes
`Inception-v3`	Inception-v3 convolutional neural network. For the pretrained Inception-v3 model, see `inceptionv3`.	Yes	Yes
`MobileNet-v2`	MobileNet-v2 convolutional neural network. For the pretrained MobileNet-v2 model, see `mobilenetv2`.	Yes	Yes
`NASNet-Large`	NASNet-Large convolutional neural network. For the pretrained NASNet-Large model, see `nasnetlarge`.	Yes	Yes
`NASNet-Mobile`	NASNet-Mobile convolutional neural network. For the pretrained NASNet-Mobile model, see `nasnetmobile`.	Yes	Yes
`ResNet`	ResNet-18, ResNet-50, and ResNet-101 convolutional neural networks. For the pretrained ResNet models, see `resnet18`, `resnet50`, and `resnet101`.	Yes	Yes
`SegNet`	Multi-class pixelwise segmentation network. For more information, see `segnetLayers`.	No	Yes
`SqueezeNet`	Small, deep neural network. For the pretrained SqeezeNet models, see `squeezenet`.	Yes	Yes
`VGG-16`	VGG-16 convolutional neural network. For the pretrained VGG-16 model, see `vgg16`.	Yes	Yes
`VGG-19`	VGG-19 convolutional neural network. For the pretrained VGG-19 model, see `vgg19`.	Yes	Yes
`Xception`	Xception convolutional neural network. For the pretrained Xception model, see `xception`.	Yes	Yes

Supported Layers

The following layers are supported for code generation by MATLAB Coder for the target deep learning libraries specified in the table.

Once you install the support package MATLAB Coder Interface for Deep Learning Libraries, you can use coder.getDeepLearningLayers to see a list of the layers supported for a specific deep learning library. For example:

coder.getDeepLearningLayers('mkldnn')

Layer Name	Description	ARM Compute Library	Intel MKL-DNN
`additionLayer`	Addition layer	Yes	Yes
`anchorBoxLayer`	Anchor box layer	Yes	Yes
`averagePooling2dLayer`	Average pooling layer	Yes	Yes
`batchNormalizationLayer`	Batch normalization layer	Yes	Yes
`bilstmLayer`	Bidirectional LSTM layer	Yes	No
`classificationLayer`	Create classification output layer	Yes	Yes
`clippedReluLayer`	Clipped Rectified Linear Unit (ReLU) layer	Yes	Yes
`concatenationLayer`	Concatenation layer	Yes	Yes
`convolution2dLayer`	2-D convolution layer	Yes	Yes
`crop2dLayer`	Layer that applies 2-D cropping to the input	Yes	Yes
`CrossChannelNormalizationLayer`	Channel-wise local response normalization layer	Yes	Yes
Custom output layers	All output layers including custom classification or regression output layers created by using `nnet.layer.ClassificationLayer` or `nnet.layer.RegressionLayer`. For an example showing how to define a custom classification output layer and specify a loss function, see Define Custom Classification Output Layer (Deep Learning Toolbox). For an example showing how to define a custom regression output layer and specify a loss function, see Define Custom Regression Output Layer (Deep Learning Toolbox).	Yes	Yes
`depthConcatenationLayer`	Depth concatenation layer	Yes	Yes
`dropoutLayer`	Dropout layer	Yes	Yes
`eluLayer`	Exponential linear unit (ELU) layer	Yes	Yes
`fullyConnectedLayer`	Fully connected layer	Yes	Yes
`globalAveragePooling2dLayer`	Global average pooling layer for spatial data	Yes	Yes
`globalMaxPooling2dLayer`	2-D global max pooling layer	Yes	Yes
`groupedConvolution2dLayer`	2-D grouped convolutional layer	Yes If you specify an integer for `numGroups`, then the value must be less than or equal to `2`.	Yes
`imageInputLayer`	Image input layer Code generation does not support `'Normalization'` specified using a function handle.	Yes	Yes
`leakyReluLayer`	Leaky Rectified Linear Unit (ReLU) layer	Yes	Yes
`lstmLayer`	Long short-term memory (LSTM) layer	Yes	No
`maxPooling2dLayer`	Max pooling layer	Yes	Yes
`maxUnpooling2dLayer`	Max unpooling layer	No	Yes
`pixelClassificationLayer`	Create pixel classification layer for semantic segmentation	Yes	Yes
`regressionLayer`	Create a regression output layer	Yes	Yes
`reluLayer`	Rectified Linear Unit (ReLU) layer	Yes	Yes
`sequenceInputLayer`	Sequence input layer	Yes	No
`softmaxLayer`	Softmax layer	Yes	Yes
`ssdMergeLayer`	SSD merge layer for object detection	Yes	Yes
`nnet.keras.layer.FlattenCStyleLayer`	Flattens activations into 1-D assuming C-style (row-major) order	Yes	Yes
`nnet.keras.layer.GlobalAveragePooling2dLayer`	Global average pooling layer for spatial data	Yes	Yes
`nnet.keras.layer.SigmoidLayer`	Sigmoid activation layer	Yes	Yes
`nnet.keras.layer.TanhLayer`	Hyperbolic tangent activation layer	Yes	Yes
`nnet.keras.layer.ZeroPadding2dLayer`	Zero padding layer for 2-D input	Yes	Yes
`nnet.onnx.layer.ElementwiseAffineLayer`	Layer that performs element-wise scaling of the input followed by an addition	Yes	Yes
`nnet.onnx.layer.FlattenLayer`	Flatten layer for ONNX™ network	Yes	Yes
`nnet.onnx.layer.IdentityLayer`	Layer that implements ONNX identity operator	Yes	Yes
`tanhLayer`	Hyperbolic tangent (tanh) layer	Yes	Yes
`transposedConv2dLayer`	Transposed 2-D convolution layer Code generation does not support asymmetric cropping of the input. For example, specifying a vector `[t b l r]` for the `'Cropping'` parameter to crop the top, bottom, left, and right of the input is not supported.	Yes	Yes
`wordEmbeddingLayer`	A word embedding layer maps word indices to vectors	Yes	No
`YOLOv2OutputLayer`	Output layer for YOLO v2 object detection network	Yes	Yes
`YOLOv2ReorgLayer`	Reorganization layer for YOLO v2 object detection network	Yes	Yes
`YOLOv2TransformLayer`	Transform layer for YOLO v2 object detection network	Yes	Yes

Supported Classes

Class	Description	ARM Compute Library	Intel MKL-DNN
`yolov2ObjectDetector`	Only the `detect` method of the `yolov2ObjectDetector` is supported for code generation. The `roi` argument to the `detect` method must be a code generation constant (`coder.const()`) and a 1x4 vector. Only the `Threshold`, `SelectStrongest`, `MinSize`, and `MaxSize` name-value pairs for `detect` are supported. The `labels` output of `detect` is returned as a cell array of character vectors, for example, `{'car','bus'}`.	Yes	Yes
`ssdObjectDetector`	Object to detect objects using the SSD-based detector. Only the `detect` method of the `ssdObjectDetector` is supported for code generation. The `roi` argument to the `detect` method must be a codegen constant (`coder.const()`) and a 1x4 vector. Only the `Threshold`, `SelectStrongest`, `MinSize`, `MaxSize`, and `MiniBatchSize` Name-Value pairs are supported. All Name-Value pairs must be compile-time constants. The channel and batch size of the input image must be fixed size. The `labels` output is returned as a categorical array. In the generated code, the input is rescaled to the size of the input layer of the network. But the bounding box that the `detect` method returns is in reference to the original input size.	Yes	Yes

Class

Description

ARM Compute Library

Intel MKL-DNN

yolov2ObjectDetector

Only the detect method of the yolov2ObjectDetector is supported for code generation.
The roi argument to the detect method must be a code generation constant (coder.const()) and a 1x4 vector.
Only the Threshold, SelectStrongest, MinSize, and MaxSize name-value pairs for detect are supported.
The labels output of detect is returned as a cell array of character vectors, for example, {'car','bus'}.

Yes

ssdObjectDetector

Object to detect objects using the SSD-based detector.

Only the detect method of the ssdObjectDetector is supported for code generation.
The roi argument to the detect method must be a codegen constant (coder.const()) and a 1x4 vector.
Only the Threshold, SelectStrongest, MinSize, MaxSize, and MiniBatchSize Name-Value pairs are supported. All Name-Value pairs must be compile-time constants.
The channel and batch size of the input image must be fixed size.
The labels output is returned as a categorical array.
In the generated code, the input is rescaled to the size of the input layer of the network. But the bounding box that the detect method returns is in reference to the original input size.

Yes

Documentation