Completed
Overview
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Scaling Training and Batch Inference - A Deep Dive into Ray AIR's Data Processing Engine
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Overview
- 3 ML Pipelines Must Scale with Data
- 4 Distributed Data-Parallel to the Rescue
- 5 Scaling the Typical ML Pipeline
- 6 Possible Solution - Coordinated Pipelining
- 7 Ray Datasets: AIR's Data Processing Engine
- 8 Avoiding GPU Data Prep Stalls
- 9 Dataset Sharding
- 10 Parallel I/O and Transformations
- 11 Dataplane Optimizations
- 12 Pipelining Ingest with Training
- 13 Pipelining Ingest with Inference
- 14 Autoscaling Actor Pool for Inference
- 15 Per-epoch Shuffling - Distributed
- 16 ML engineer at Telematics Startup
- 17 Summary
