Click on the button below to download the materials.
The demand for AI applications across industries emphasizes the necessity for effective deployment. This is especially the case for engineering solutions that run on edge nodes or embedded processors. To guarantee a performant operation, these setups often rely on GPU accelerated hardware making the deployment more challenging.
In this workshop, participants learn workflows for developing and deploying their AI models for GPU accelerated hardware such as NVIDIA Jetson boards. The workflow is highlighted with the implementation of an AI based application for signal processing. Throughout the workshop, participants will:
-
Design and train a neural network and integrate it into a simulation environment to perform functional validation.
-
Learn options how to import already trained networks from open-source frameworks into MATLAB.
-
Generate optimized embedded C code from the AI model, analyze performance and validate functionality using Software-in-the-Loop and Processor-in-the-Loop simulation.
-
Learn how to deploy the final application to an embedded device.
By attending this workshop, participants will acquire valuable knowledge and hands-on experience in developing and deploying AI applications with NVIDIA GPUs.
- Basic understanding of machine learning and deep learning concepts
- Programming experience - MATLAB experience is helpful.
- Participants that would like to get familiar with MATLAB and Simulink before the workshop are encouraged to check out our free onramp trainings under matlabacademy.mathworks.com)
- MathWorks account (required to access MATLAB Online and the course content)
- Laptop with internet access and browser (Edge or Chrome recommended)