The Issue 26 of Intel Parallel Universe Magazine has a number of interesting materials.
Intel Python team heard your feedback to provide more detailed use case how various technologies being shipped within Intel(R) Distribution for Python* can be applied to accelerate numerical and machine learning applications. Together with experts from Intel(R) Data Analytics Acceleration Library and Intel(R) VTune(TM) Amplifier teams we created a detailed case study showing how to accelerate one of typical machine learning problems, the Recommendation System. In our step-by-step study we illustrate the value of these technologies and give general recommendations how to get speeds near to native code performance. Read the article Overcome Python* Performance Barriers for Machine Learning.
I also recommend reading materials from Intel MKL and DAAL library experts,Unleash the Power of Big Data Analytics and Machine Learning, Lightning-Fast R* Machine Learning Algorithms, A Performance Library for Data Analytics and Machine Learning, and MeritData Speeds Up Its Tempo* Big Data Platform Using Intel® High-Performance Libraries to get hands-on experience with machine learning capabilities of these products.
Don't miss to read about next Intel VTune Amplifier capabilities to seamlessly profile Python and Java codes while working on your Big Data and Machine Learning projects in this article Profililing Java* and Python* Code using Intel® VTune™ Amplifier.
The last but not least, for C/C++ and Fortran worlds, learn more about optimizing Intel(R) Xeon(TM) Phi using world-class compilers and libraries from Intel(R) Parallel Studio XE 2018 Modernize Your Code for Intel® Xeon Phi™ Processors.
Get the full issue here.
Thank you,
Sergey Maidanov
Software Engineering Manager for Intel Distribution for Python
