The Painless Route in Python to Fast and Scalable Machine Learning

Discover a painless route to fast and scalable machine learning in Python through this 43-minute conference talk from EuroPython 2020. Learn how to achieve close-to-native performance using Intel-optimized packages like numpy, scipy, and scikit-learn. Explore techniques for high performance and scalability across multi-core machines and large clusters. Gain insights into using Scalable Dataframe Compiler (SDC) for efficient processing of huge datasets with pandas/Python, and leverage daal4py's convenient Python API for scalable machine learning algorithms. Understand how to combine SDC and daal4py to build end-to-end analytics pipelines that scale to clusters with minimal code changes, achieving performance comparable to hand-tuned C++/MPI code.

V. Fedotova, F. Schlimbach - The Painless Route in Python to Fast and Scalable Machine Learning