Histogram-based Gradient Boosting in scikit-learn 0.21
Offered By: EuroPython Conference via YouTube
Course Description
Overview
Discover the latest advancements in scikit-learn 0.21 through this 45-minute EuroPython Conference talk. Explore the new implementation of Gradient Boosted Trees, focusing on the histogram-based approach for evaluating tree node split candidates. Learn how this update improves computational performance, making it competitive with specialized libraries like XGBoost and LightGBM. Gain insights into the efficient use of multi-core CPUs and compare the numba-based prototype (pygbm) with the final cython implementation. Understand the developer experience differences between numba and cython in the context of machine learning model development.
Syllabus
Olivier Grisel - Histogram-based Gradient Boosting in scikit-learn 0.21
Taught by
EuroPython Conference
Related Courses
Machine LearningUniversity of Washington via Coursera Machine Learning
Stanford University via Coursera Machine Learning
Georgia Institute of Technology via Udacity Statistical Learning with R
Stanford University via edX Machine Learning 1—Supervised Learning
Brown University via Udacity