To achieve substantial speed-ups, take advantage of key Intel® architectural innovations and optimized data-science and machine learning libraries via a smart application of NumPy, SciPy, and pandas techniques.
In this session:
- Learn about NumPy aggregations, universal functions, broadcasting, and other techniques used to expose how CPU vectorization works.
- Use these techniques to achieve outsized performance gains by replacing Python* loop-centric or list-comprehension applications with smarter equivalents that are more maintainable, more efficient, and much faster.
- Measure your code’s acceleration to discover achieved performance boost, whether that’s 10x or over 100x.