AI on Intel Architecture

This session offers an overview of the hardware and software which empower AI on Intel.
First, we will introduce the hardware features used for AI on Intel. We will then get a first glance at the software stack harnessing them, namely the Intel® oneAPI AI Analytics Toolkit.

Next, we will cover the Intel-optimized libraries for Machine Learning: Modin, Intel® Extension for Scikit-learn and XGBoost. Also we will look at the optimizations of Deep Learning frameworks such as Intel® Extension for PyTorch (IPEX) and Tensorflow (ITEX) powered by highly-optimized Intel® oneDNN. You will learn how to use the Intel extensions to enable extra features such as mixed-precision and running Deep Learning computations on Intel dGPU.

Last, We will also present the Intel Neural Compressor (i.e., a comprehensive software toolkit that offers a plethora of optimization for neural networks, such as quantization, pruning, graph optimizations.)

Download Presentation

×


Register for the oneAPI DevSummit hosted by UXL:

Register Now