Introduction to Intel® Distribution of OpenVINO™ Toolkit

Course Description

This course introduces the AI algorithms and framework in the Intel® Distribution of OpenVINO™ toolkit, which is used to solve complex problems.

  • This toolkit is a suite of tools for performing optimizations and inference on trained deep learning models into Python*, C, and C++ applications, and deploying these applications to the edge, network edge, or cloud.
  • It provides acceleration on Intel® CPUs, GPUs, VPUs, and other Intel® hardware architecture accelerators.
  • The OpenVINO toolkit is compatible with many common libraries such as TensorFlow*, PyTorch*, ONNX* (Open Neural Network Exchange), and many others, and provides increased performance above stock libraries on Intel® architecture.

Included in this course:

  • 5 modules (Estimated time to complete: 31 hours)
  • 9 lab exercises

 

Modules
  • Introduction to AI and the Intel Distribution of OpenVINO Toolkit
  • Optimization and Quantization of AI Models for Improved Performance
  • Create Scalable and Future-Ready AI Applications with the Inference Engine
  • Hardware Accelerators for Deep Learning
  • Streamline AI Application Development with the Deep Learning Workbench  
Get the Assignments and Quizzes

Five modules and nine lab exercises with slide presentations and quizzes are available as a separate download.

Details

Learning Objectives

After completing this course, students will be able to:

  • Analyze and optimize deep learning models for computer vision, natural language processing, and more.
  • Describe and program the OpenVINO API into applications to run deep learning inference.
  • Deploy inference for deep learning models heterogeneously.
  • Describe where to download and install the OpenVINO toolkit.

Target Audience

Senior undergraduate and graduate students studying:

  • Computer science
  • Engineering
  • Science and mathematics

Prerequisites

  • Python* programming at a minimum; C or C++ programming is beneficial
  • Basic understanding of neural network trained models, and their weights and biases