Is it possible to create VHDL Code of the “Image Classifier” block from the Deep Neural Networks library with HDL Coder?

I have created a neural network to classify digits. With the “Image Classifier” block from the Deep Neural Networks library I built a Simulink model, which worked well. Now I tried to generate the VHDL code of these NN using the HDL Coder. I get following error:

Unable to resolve the name coder.loadMatObj.

Function 'Subsystem/Image Classifier/MLFB' (#178.49.218), line 2, column 14: "deep.blocks.internal.classifier(image, 'Ziffernerkennung2/Subsystem/Image Classi"

Is it possible to create VHDL code of the “Image Classifier” block from the Deep Neural Networks library with HDL Coder? Do I need an extra Toolbox?

Best regards,

Martin Berthold

Answer 1

[Siyuan Xu]

Hi,

Regarding generating HDL you can refer to Deep Learning HDL Toolbox & Supported Networks, Layers and Boards.

It allows you to run a variety of deep learning networks on supported Xilinx® and Intel® FPGA and SoC devices.

Best regards,

Siyuan