Developing Xilinx AI Solutions for Cloud-based Applications

This course describes how to use Xilinx machine learning solutions for data center and cloud-based applications.

The emphasis of this course is on:

  • Utilizing DNN algorithms, models, inference and training, and frameworks in the cloud
  • Using the xfDNN software stack to optimize the trained model
  • Optimizing the xDNN instructions for the xDNN processing engine in a cloud application

You can also take the full assessment for the entire course.

1 Introduction to Xilinx Machine Learning Solutions for Cloud and Edge Applications
Describes Xilinx machine learning solutions.
2 Overview of ML Concepts
Overview of ML concepts such as DNN algorithms, models, inference and training, and frameworks.
3 xDNN Architecture Overview
Describes the architectural features of the Xilinx Deep Neural Network (xDNN) processing engine and how the engine can be optimized for cloud applications.
4 xfDNN Middleware
Describes the xfDNN middleware, a high-performance software library with a well-defined API, which acts as a bridge between deep learning frameworks such as Caffe, MXNet, TensorFlow, and the xDNN IP running on an FPGA
5 ML Suite-Supported Frameworks
Describes the support for many common machine learning frameworks such as Caffe, MXNet, and TensorFlow as well as Python and RESTful APIs.

Using ML Suite for Custom Applications with the Alveo Card
Describes the environment to set up the Xilinx ML Suite and how to create a custom application and deploy the design.

You can also view a video demonstration illustrating the above concepts.