Deep Learning Support from MATLAB Coder
Capabilities and Features
MATLAB Coder generates C and C++ code from MATLAB code for a variety of hardware platforms, from desktop systems to embedded hardware. It supports most of the MATLAB language and a wide range of toolboxes. You can integrate the generated code into your projects as source code, static libraries, or dynamic libraries. The generated code is readable and portable.
You can deploy a variety of trained deep learning networks, such as YOLO, ResNet-50, SegNet, and MobileNet, from Deep Learning Toolbox to NVIDIA GPUs. You can generate optimized code for preprocessing and postprocessing along with your trained deep learning networks to deploy complete algorithms.
MATLAB Coder Interface for Deep Learning Libraries provides the ability to customize the generated code from deep learning algorithms by leveraging target specific acceleration libraries on the embedded target. With this support package, you can integrate with libraries optimized for specific CPU targets for deep learning such as the ARM® Compute Library for ARM architectures.
MATLAB Coder Interface for Deep Learning integrates with the following deep learning accelerator libraries and the corresponding CPU architectures:
- Intel® Math Kernel Library for Deep Neural Networks (MKL-DNN) for Intel CPUs that support AVX2
- The generated code leverages the Intel MKL-DNN library, which is an open source performance library for deep learning applications, providing vectorized and threaded building blocks optimized for Intel architectures.
- ARM Compute library for ARM Cortex® A family of processors that support NEON™ instructions
- The generated code leverages the ARM compute library, which is a collection of low-level software functions optimized for certain ARM architectures, targeting image processing, computer vision, and machine learning applications.
Platform and Release Support
Available on 64-bit Microsoft and 64-bit Ubuntu only.
See the hardware support package system requirements table for current and prior version, release, and platform availability.