MLlib Strata - UC Berkeley AMP Camp PDF
Preview MLlib Strata - UC Berkeley AMP Camp
ML base ML base ML base ML base ML base MLlib & ML base ML base ML base Distributed Machine Learning on ML base Evan Sparks UC Berkeley January 31st, 2014 Collaborators: Ameet Talwalkar, Xiangrui Meng, Virginia Smith, Xinghao Pan, Shivaram Venkataraman, Matei Zaharia, Rean Griffith, John Duchi, Joseph Gonzalez, Michael Franklin, Michael I. Jordan, Tim Kraska www.mlbase.org UC Berkeley Declarative result ML Task (e.g., fn-model & summary) User Master Server Problem: Scalable implementa.ons Parser Meta-Data difficult for ML Developers… ML Contract + Code LLP M ML Library a s t e Optimizer r Statistics PLP Executor/Monitoring ML Developer DMX DMX DMX DMX Runtime Runtime Runtime Runtime S …. l a v e s Declarative result ML Task (e.g., fn-model & summary) User Master Server Problem: Scalable implementa.ons Parser Meta-Data difficult for ML Developers… ML Contract + Code LLP M ML Library a s t e Optimizer r Statistics PLP Executor/Monitoring ML Developer DMX DMX DMX DMX Runtime Runtime Runtime Runtime S …. l a v e s Declarative result ML Task (e.g., fn-model & summary) User Master Server Problem: Scalable implementa.ons Parser Meta-Data difficult for ML Developers… ML Contract + Code LLP M ML Library a s t e Optimizer r Statistics PLP Executor/Monitoring ML Developer DMX DMX DMX DMX Runtime Runtime Runtime Runtime S …. l a v e s Problem: ML is difficult for End Users… Too many algorithms… Problem: ML is difficult for End Users… Too many knobs… Too many algorithms… Problem: ML is difficult for End Users… Too many knobs… Too many algorithms… Difficult to debug… Problem: ML is difficult for End Users… Too many knobs… Too many algorithms… Difficult to debug… Doesn’t scale… Problem: ML is difficult t s a F R e l for End Users… i a b l e e Accurate l b a v Too many o r P knobs… Too many algorithms… Difficult to debug… Doesn’t scale… ML Experts MLbase Systems Experts
Description:The list of books you might like
