Building Deep Learning Applications Using PyTorch on AWS
Are you looking to accelerate going from prototyping to production for your PyTorch-based models? AWS offers services to build and scale ML models using Amazon SageMaker for a fully managed experience, Amazon Elastic Kubernetes Service (EKS) for containerized workloads, and AWS ParallelCluster to run Amazon EC2 clusters to easily train and deploy your workloads.
Distributed ML training using PyTorch on High-Performance Computing (HPC) clusters
June 29, 7:00 AM - 8:30 AM PDT | 10:00 AM - 11:30 AM EDT
By using PyTorch Fully Sharded Data Parallel (FSDP) library for distributed training with powerful Amazon EC2 instances and AWS ParallelCluster, you can easily implement distributed training architectures to accelerate training for large ML models. Attend this hands-on workshop to learn about best practices when deploying distributed training architectures on AWS using EC2 and PyTorch FSDP library.
Learning objective:
- Learn how to get started with AWS on PyTorch
- Learn how to create a distributed training architecture using AWS ParallelCluster
- Learn about PyTorch Fully Sharded Data Parallel (FSDP) library for distributed training
Who should attend:
Data Scientists, Developers, ML Practitioners, MLOps, and Software Engineers
Speaker and Presenter Information
Pierre Yves, Specialist SA Manager, Frameworks AI/ML, AWS
Shubha Kumbadakone, Specialist BD, Frameworks AI/ML, AWS
Shashank Prasanna, Sr. Developer Advocate AI/ML, AWS
Relevant Government Agencies
Other Federal Agencies, Federal Government, State & Local Government
Event Type
Webcast
This event has no exhibitor/sponsor opportunities
When
Wed, Jun 29, 2022, 7:00am - 9:00am
PT
Cost
Complimentary: $ 0.00
Website
Click here to visit event website
Organizer
Amazon Web Services (AWS)