Tips and Tricks for Running Machine Learning at Scale with Vertica (Boston)
Join this session to learn then adopt the most popular and helpful machine learning techniques in Vertica. Hear directly from the inventor of VerticaPy, a powerful Python library that exposes scikit-like functionality to conduct data science projects on data stored in Vertica. VerticaPy has many features that will help you achieve high performance at scale without impacting your cluster. Topics covered include: caching some statistics to avoid recomputing them twice, sending multiple queries iteratively to preserve your cluster, and even sending multiple queries at the same time to gain performance.