{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/fastbdt-a-speed-optimized-and-cache-friendly","title":"FastBDT: A speed-optimized and cache-friendly implementation of stochastic gradient-boosted decision trees for multivariate classification","arxiv_id":"1609.06119","date":"2016-09-20","proceeding":null,"authors":["Thomas Keck"],"abstract":"Stochastic gradient-boosted decision trees are widely employed for\nmultivariate classification and regression tasks. This paper presents a\nspeed-optimized and cache-friendly implementation for multivariate\nclassification called FastBDT. FastBDT is one order of magnitude faster during\nthe fitting-phase and application-phase, in comparison with popular\nimplementations in software frameworks like TMVA, scikit-learn and XGBoost. The\nconcepts used to optimize the execution time and performance studies are\ndiscussed in detail in this paper. The key ideas include: An equal-frequency\nbinning on the input data, which allows replacing expensive floating-point with\ninteger operations, while at the same time increasing the quality of the\nclassification; a cache-friendly linear access pattern to the input data, in\ncontrast to usual implementations, which exhibit a random access pattern.\nFastBDT provides interfaces to C/C++, Python and TMVA. It is extensively used\nin the field of high energy physics by the Belle II experiment.","url_abs":"http://arxiv.org/abs/1609.06119v1","url_pdf":"http://arxiv.org/pdf/1609.06119v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"fastbdt-a-speed-optimized-and-cache-friendly","repo_url":"https://github.com/thomaskeck/FastBDT","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}