Supported Networks, Layers, Boards, and Tools

Supported Pretrained Networks

Deep Learning HDL Toolbox™ supports code generation for series convolutional neural networks (CNNs or ConvNets). You can generate code for any trained convolutional neural network whose computational layers are supported for code generation. See Supported Layers. You can use one of the pretrained networks listed in the table and generate code for your target Intel^® or Xilinx^® FPGA boards.

Network	Network Description	Type	Single Data Type (with Shipping Bitstreams)			INT8 data type (with Shipping Bitstreams)			Application Area
			ZCU102	ZC706	Arria10 SoC	ZCU102	ZC706	Arria10 SoC	Classification
AlexNet	AlexNet convolutional neural network.	Series Network	Yes	Yes	Yes	Yes	Yes	Yes	Classification
LogoNet	Logo recognition network (LogoNet) is a MATLAB^® developed logo identification network. For more information, see Logo Recognition Network.	Series Network	Yes	Yes	Yes	Yes	Yes	Yes	Classification
MNIST	MNIST Digit Classification. See Create Simple Deep Learning Network for Classification	Series Network	Yes	Yes	Yes	Yes	Yes	Yes	Classification
Lane detection	LaneNet convolutional neural network. For more information, see Deploy Transfer Learning Network for Lane Detection	Series Network	Yes	Yes	Yes	Yes	Yes	Yes	Classification
VGG-16	VGG-16 convolutional neural network. For the pretrained VGG-16 model, see `vgg16`.	Series Network	No. Network exceeds PL DDR memory size	No. Network exceeds FC module memory size.	Yes	Yes	No. Network exceeds FC module memory size.	Yes	Classification
VGG-19	VGG-19 convolutional neural network. For the pretrained VGG-19 model, see `vgg19` .	Series Network	No. Network exceeds PL DDR memory size	No. Network exceeds FC module memory size.	Yes	Yes	No. Network exceeds FC module memory size.	Yes	Classification
Darknet-19	Darknet-19 convolutional neural network. For the pretrained darknet-19 model, see `darknet19`.	Series Network	Yes	Yes	Yes	Yes	Yes	Yes	Classification
Radar Classification	Convolutional neural network that uses micro-Doppler signatures to identify and classify the object. For more information, see Bicyclist and Pedestrian Classification by Using FPGA.	Series Network	Yes	Yes	Yes	Yes	Yes	Yes	Classification and Software Defined Radio (SDR)
Defect Detection `snet_defnet`	`snet_defnet` is a custom AlexNet network used to identify and classify defects. For more information, see Defect Detection.	Series Network	Yes	Yes	Yes	Yes	Yes	Yes	Classification
Defect Detection `snet_blemdetnet`	`snet_blemdetnet` is a custom convolutional neural network used to identify and classify defects. For more information, see Defect Detection.	Series Network	Yes	Yes	Yes	Yes	Yes	Yes	Classification
YOLO v2 Vehicle Detection	You look only once (YOLO) is an object detector that decodes the predictions from a convolutional neural network and generates bounding boxes around the objects. For more information, see Vehicle Detection Using YOLO v2 Deployed to FPGA	Series Network based	Yes	Yes	Yes	Yes	Yes	Yes	Object detection
DarkNet-53	Darknet-53 convolutional neural network. For the pretrained DarkNet-53 model, see `darknet53`.	Directed acyclic graph (DAG) network based	No. Network exceeds PL DDR memory size.	No. Network exceeds PL DDR memory size.	Yes	Yes	Yes	Yes	Classification
ResNet-18	ResNet-18 convolutional neural network. For the pretrained ResNet-18 model, see `resnet18`.	Directed acyclic graph (DAG) network based	Yes		Yes	Yes	Yes	Yes	Classification
ResNet-50	ResNet-50 convolutional neural network. For the pretrained ResNet-50 model, see `resnet50`.	Directed acyclic graph (DAG) network based	No. Network exceeds PL DDR memory size.	No. Network exceeds PL DDR memory size.	Yes	Yes	Yes	Yes	Classification
ResNet-based YOLO v2	You look only once (YOLO) is an object detector that decodes the predictions from a convolutional neural network and generates bounding boxes around the objects. For more information, see Vehicle Detection Using DAG Network Based YOLO v2 Deployed to FPGA.	Directed acyclic graph (DAG) network based	Yes	Yes	Yes	Yes	Yes	Yes	Object detection
MobileNetV2	MobileNet-v2 convolutional neural network. For the pretrained MobileNet-v2 model, see `mobilenetv2`.	Directed acyclic graph (DAG) network based	Yes	Yes	Yes	No	No	No	Classification

Supported Layers

The following layers are supported by Deep Learning HDL Toolbox.

Input Layers

Layer	Layer Type Hardware (HW) or Software(SW)	Description and Limitations	INT8 Compatible
`imageInputLayer`	SW	An image input layer inputs 2-D images to a network and applies data normalization.	Yes. Runs as single datatype in SW.

Convolution and Fully Connected Layers

Layer	Layer Type Hardware (HW) or Software(SW)	Layer Output Format	Description and Limitations	INT8 Compatible
`convolution2dLayer`	HW	Convolution (Conv)	A 2-D convolutional layer applies sliding convolutional filters to the input. These limitations apply when generating code for a network using this layer: Filter size must be 1-14 and square. For example [1 1] or [14 14]. Stride size must be 1-12 and square. Padding size must be in the range 0-8. Dilation factor must be [1 1]. Padding value is not supported.	Yes
`groupedConvolution2dLayer`	HW	Convolution (Conv)	A 2-D grouped convolutional layer separates the input channels into groups and applies sliding convolutional filters. Use grouped convolutional layers for channel-wise separable (also known as depth-wise separable) convolution. Code generation is now supported for a 2-D grouped convolution layer that has the `NumGroups` property set as `'channel-wise'`. These limitations apply when generating code for a network using this layer: Filter size must be 1-14 and square. For example [1 1] or [14 14]. When the `NumGroups` is set as `'channel-wise'`, filter size must be 3-14. Stride size must be 1-12 and square. Padding size must be in the range 0-8. Dilation factor must be [1 1]. Number of groups must be 1 or 2.	Yes
`fullyConnectedLayer`	HW	Fully Connected (FC)	A fully connected layer multiplies the input by a weight matrix, and then adds a bias vector. These limitations apply when generating code for a network using this layer: The layer input and output size are limited by the values specified in InputMemorySize and OutputMemorySize.	Yes

Activation Layers

Layer	Layer Type Hardware (HW) or Software(SW)	Layer Output Format	Description and Limitations	INT8 Compatible
`reluLayer`	HW	Layer is fused.	A ReLU layer performs a threshold operation to each element of the input where any value less than zero is set to zero. A ReLU layer is supported only when it is preceded by any of these layers: convolution layer fully connected layer adder layer	Yes
`leakyReluLayer`	HW	Layer is fused.	A leaky ReLU layer performs a threshold operation where any input value less than zero is multiplied by a fixed scalar. A leaky ReLU layer is supported only when it is preceded by any of these layers: convolution layer fully connected layer adder layer	Yes
`clippedReluLayer`	HW	Layer is fused.	A clipped ReLU layer performs a threshold operation where any input value less than zero is set to zero and any value above the clipping ceiling is set to that clipping ceiling value. A clipped ReLU layer is supported only when it is preceded by any of these layers: convolution layer fully connected layer adder layer	Yes

Normalization, Dropout, and Cropping Layers

Layer	Layer Type Hardware (HW) or Software(SW)	Layer Output Format	Description and Limitations	INT8 Compatible
`batchNormalizationLayer`	HW	Layer is fused.	A batch normalization layer normalizes each input channel across a mini-batch. A batch normalization layer is only supported only when it is preceded by a convolution layer.	Yes
`crossChannelNormalizationLayer`	HW	Convolution (Conv)	A channel-wise local response (cross-channel) normalization layer carries out channel-wise normalization. The `WindowChannelSize` must be in the range of 3-9 for code generation.	Yes. Runs as single datatype in HW.
`dropoutLayer`	NoOP on inference	NoOP on inference	A dropout layer randomly sets input elements to zero with a given probability.	Yes

Pooling and Unpooling Layers

Layer	Layer Type Hardware (HW) or Software(SW)	Layer Output Format	Description and Limitations	INT8 Compatible
`maxPooling2dLayer`	HW	Convolution (Conv)	A max pooling layer performs down sampling by dividing the input into rectangular pooling regions and computing the maximum of each region. These limitations apply when generating code for a network using this layer: Pool size must be 1-14 and square. For example [1 1] or [12 12]. Stride size must be 1-7 and square. Padding size must be in the range 0-2.	Yes
`averagePooling2dLayer`	HW	Convolution (Conv)	An average pooling layer performs down sampling by dividing the input into rectangular pooling regions and computing the average values of each region. These limitations apply when generating code for a network using this layer: Pool size must be 1-14 and square. For example [3 3] Stride size must be 1-7 and square. Padding size must be in the range 0-2.	Yes
`globalAveragePooling2dLayer`	HW	Convolution (Conv) or Fully Connected (FC). When the input activation size is lower than the memory threshold, the layer output format is FC.	A global average pooling layer performs down sampling by computing the mean of the height and width dimensions of the input. These limitations apply when generating code for a network using this layer: Pool size value must be in the range 1-14 for a conv module based global average pooling layer and in the range 1-12 for an FC based global average pooling layer. The pool size value must be square. For example, [1 1] or [12 12]. Total activation pixel size must be smaller than the deep learning processor convolution module input memory size. For more information, see InputMemorySize	Yes

Combination Layers

Layer Layer Type Hardware (HW) or Software(SW) Layer Output Format Description and Limitations INT8 Compatible

additionLayer

Inherit from input.

An addition layer adds inputs from multiple neural network layers element-wise.

These limitations apply when generating code for a network using this layer:

The maximum number of inputs to the addition layer is two when the input data type is int8.
Both input layers must have the same output layer format. For example, both layers must have conv output format or fc output format.

Yes

depthConcatenationLayer

Inherit from input.

A depth concatenation layer takes inputs that have the same height and width and concatenates them along the third dimension (the channel dimension).

These limitations apply when generating code for a network using this layer:

The input activation feature number must be a multiple of the square root of the ConvThreadNumber.
Inputs to the depth concatenation layer must be exclusive to the depth concatenation layer.
Layers that have a conv output format and layers that have an FC output format cannot be concatenated together.

Output Layer

Layer	Layer Type Hardware (HW) or Software(SW)	Description and Limitations	INT8 Compatible
`softmax`	SW	A softmax layer applies a softmax function to the input.	Yes. Runs as single datatype in SW.
`classificationLayer`	SW	A classification layer computes the cross-entropy loss for multi class classification issues with mutually exclusive classes.	Yes
`regressionLayer`	SW	A regression layer computes the half-mean-squared-error loss for regression problems.	Yes