filmov
tv
Ray: A Framework for Scaling and Distributing Python & ML Applications
Показать описание
Recording of a live meetup on Feb 16, 2022 from our friends at Data + AI Denver/Boulder meetup group.
Meetup details:
Our first talk of the year features Jules Damji, Lead Developer Advocate at Anyscale as he discusses Ray: A Framework for Scaling and Distributing Python & ML Applications.
ABOUT THE TALK:
Modern machine learning (ML) workloads, such as deep learning and large-scale model training, are compute-intensive and require distributed execution. Ray is an open-source, distributed framework from U.C. Berkeley’s RISELab that easily scales Python applications and ML workloads from a laptop to a cluster, with an emphasis on the unique performance challenges of ML/AI systems. It is now used in many production deployments.
This talk will cover Ray’s overview, architecture, core concepts, and primitives, such as remote Tasks and Actors; briefly discuss Ray native libraries (Ray Tune, Ray Train, Ray Serve, Ray Datasets, RLlib); and Ray’s growing ecosystem.
Through a demo using XGBoost for classification, we will demonstrate how you can scale training, hyperparameter tuning, and inference—from a single node to a cluster, with tangible performance difference when using Ray.
The takeaways from this talk are:
Learn Ray architecture, core concepts, and Ray primitives and patterns
Why Distributed computing will be the norm not an exception
How to scale your ML workloads with Ray libraries:
Training on a single node vs. Ray cluster, using XGBoost with/without Ray
Hyperparameter search and tuning, using XGBoost with Ray and Ray Tune
Inferencing at scale, using XGBoost with/without Ray
ABOUT OUR SPEAKER:
Our Speaker, Jules Damji is the Lead Developer Advocate at Anyscale Inc.
He is an MLflow contributor, and co-author of Learning Spark, 2nd Edition. He is a hands-on developer with over 25 years of experience and has worked at leading companies, such as Sun Microsystems, Netscape, @Home, Opsware/LoudCloud, VeriSign, ProQuest, Hortonworks, and Databricks, building large-scale distributed systems. He holds a B.Sc and M.Sc in computer science (from Oregon State University and Cal State, Chico respectively), and an MA in political advocacy and communication (from Johns Hopkins University).
Meetup details:
Our first talk of the year features Jules Damji, Lead Developer Advocate at Anyscale as he discusses Ray: A Framework for Scaling and Distributing Python & ML Applications.
ABOUT THE TALK:
Modern machine learning (ML) workloads, such as deep learning and large-scale model training, are compute-intensive and require distributed execution. Ray is an open-source, distributed framework from U.C. Berkeley’s RISELab that easily scales Python applications and ML workloads from a laptop to a cluster, with an emphasis on the unique performance challenges of ML/AI systems. It is now used in many production deployments.
This talk will cover Ray’s overview, architecture, core concepts, and primitives, such as remote Tasks and Actors; briefly discuss Ray native libraries (Ray Tune, Ray Train, Ray Serve, Ray Datasets, RLlib); and Ray’s growing ecosystem.
Through a demo using XGBoost for classification, we will demonstrate how you can scale training, hyperparameter tuning, and inference—from a single node to a cluster, with tangible performance difference when using Ray.
The takeaways from this talk are:
Learn Ray architecture, core concepts, and Ray primitives and patterns
Why Distributed computing will be the norm not an exception
How to scale your ML workloads with Ray libraries:
Training on a single node vs. Ray cluster, using XGBoost with/without Ray
Hyperparameter search and tuning, using XGBoost with Ray and Ray Tune
Inferencing at scale, using XGBoost with/without Ray
ABOUT OUR SPEAKER:
Our Speaker, Jules Damji is the Lead Developer Advocate at Anyscale Inc.
He is an MLflow contributor, and co-author of Learning Spark, 2nd Edition. He is a hands-on developer with over 25 years of experience and has worked at leading companies, such as Sun Microsystems, Netscape, @Home, Opsware/LoudCloud, VeriSign, ProQuest, Hortonworks, and Databricks, building large-scale distributed systems. He holds a B.Sc and M.Sc in computer science (from Oregon State University and Cal State, Chico respectively), and an MA in political advocacy and communication (from Johns Hopkins University).
Ray: A Framework for Scaling and Distributing Python & ML Applications
0:35:23
Ray A Framework for Scaling and Distributing Python & ML Applications | Anyscale
0:15:13
Introduction to Distributed Computing with the Ray Framework
0:26:29
Ray: A Distributed Execution Framework for AI | SciPy 2018 | Robert Nishihara
0:09:44
Building Scalable Reinforcement Learning Applications with RAY Framework and Python
0:23:03
Introducing Ray Serve: Scalable and Programmable ML Serving Framework - Simon Mo, Anyscale
0:16:47
Keynote: Ray: A Distributed Framework for Heterogeneous Computing - Ion Stoica, UC Berkeley
1:01:51
Instacart Engineering Presents: Ray - A General Framework for ML and Distributed Computing
0:05:33
Xray Test Management Tool
0:34:48
Expert Talk on Scaling ML Workloads with the Ray framework by Mr. Naveen Rajan
0:26:05
ApacheConAsia2021 Keynote: Ion Stoica - Ray A universal framework for distributed computing
0:25:25
Philipp Moritz, UC Berkeley -- Ray: A Distributed Framework for Emerging AI Applications
0:03:13
Ray: A Distributed Execution Framework for Emerging AI Applications Michael Jordan (UC Berkeley)
1:02:34
Ion Stoica – Ray: A Universal Framework for Distributed Systems
0:31:17
RISE Camp 2018 03 - Ray: Distributed Execution Framework for Emerging AI, Robert Nishihara
0:27:58
Ray.jl: Julia runtime and client for the Ray compute framework | Vogt, Kleinschmidt, Moynihan
0:18:07
CodeFlare: A New Open-Source Framework For Big Data Integration And Scaling
0:35:40
Scaling Deep Learning Frameworks
0:07:06
Which Reinforcement Learning Framework is the Best?
0:33:52
The non-stationary turbulent energy cascade in the framework of scaling symmetry approach
0:13:00
Deep Learning Frameworks
0:21:37
Arras, MoonRay’s Distributed Computational Framework - Mark Jackels, DreamWorks
0:30:03
Super Fast Ray Casting in Tiled Worlds using DDA
0:03:19
Introducing fVDB: Deep Learning Framework for Generative Physical AI with Spatial Intelligence
Комментарии