Meta and AWS Collaborate on PyTorch Enhancements for AWS Trainium | Amazon Web Services

Опубликовано 21 декабря 2022, 18:23

Hear from Aparna Ramani, Vice President of AI, Data & Developer Infrastructure Engineering at Meta and PyTorch Foundation Board Member, about Meta’s work together with AWS on PyTorch enhancements for AWS purpose-built ML chip Trainium.

The enhancements include native PyTorch support for Trainium powered Trn1 instances that enables developers to integrate PyTorch with their applications and a new XLA backend to TorchDistributed that makes it easy to migrate models to Trainium instances and leverage the speed of distributed training libraries and models.

Learn more about Amazon EC2 Trn1 Instances at go.aws/3YHrZa7.

Subscribe:
More AWS videos - bit.ly/2O3zS75
More AWS events videos - bit.ly/316g9t4

ABOUT AWS
Amazon Web Services (AWS) is the world’s most comprehensive and broadly adopted cloud platform, offering over 200 fully featured services from data centers globally. Millions of customers — including the fastest-growing startups, largest enterprises, and leading government agencies — are using AWS to lower costs, become more agile, and innovate faster.

#PyTorch #Meta #AmazonSageMaker #AWS #AmazonWebServices #CloudComputing