Maximize Your CPU Resources for XGBoost* Training and Inference

@IntelDevTools

 

Gradient boosting has many real-world applications as a general-purpose machine learning technique for regression, classification, and page ranking problems. It’s a common choice for large problem sizes, yet training implementation of this method is quite complex because of the multiple kernel dependencies that impact execution time, irregular memory access, and many other issues.

If this resonates with you, register for this session to learn about Intel’s optimizations for XGBoost, with specific focus on:

  • How to speed up your boosting algorithm workloads with the Intel® oneAPI AI Analytics Toolkit
  • Example training workloads that compare the performance of the latest XGBoost implementation on an end-to-end pipeline

Your hosts are Intel AI technical engineers Mecit Gungor and Rachel Oberman.

Watch the following demos:


Get the Software

Get the Intel® oneAPI AI Analytics Toolkit, which features six powerful tools and frameworks for numerical, scientific, and machine learning applications.


Other Resources

  • Sign up for an Intel® DevCloud for oneAPI account—a free development sandbox with access to the latest Intel® hardware and oneAPI software.
  • Explore oneAPI, including developer opportunities and benefits
  • Subscribe to the POD—Code Together is an interview series that explores the challenges at the forefront of cross-architecture development. Each biweekly episode features industry VIPs who are blazing new trails through today’s data-centric world. Available wherever you get your podcasts.

Mecit Gungor
AI technical consulting engineer, Intel Corporation

Mecit is an AI technical consulting engineer and technical lead for oneAPI Technology Partner Training Program at Intel. He works on various Intel technologies that pertain to machine learning and artificial intelligence, and helps customers and partners use these technologies in their workloads. He holds a Master’s degree from Purdue, a bachelor's degree in electronics engineering, and a minor degree in mathematics from City University of Hong Kong, along with the S. H. Ho Foundation Academic achievement reward.

 

Rachel Oberman
AI technical consulting engineer, Intel Corporation

Rachel helps customers optimize their workflows with data analytics and machine learning algorithms from Intel. Prior to joining Intel in 2019, she focused on geospatial analysis and data science, and founded geoLab—an undergraduate research lab, serving as its director.

Rachel holds a bachelor’s degree in computer science and data science from the College of William & Mary in Virginia.

 

Intel® oneAPI AI Analytics Toolkit

Accelerate end-to-end machine learning and data science pipelines with optimized deep learning frameworks and high-performing Python* libraries.

Download    

See All Tools