Posted On: May 31, 2018
HAQM SageMaker now comes pre-configured to run Chainer in a Docker container, adding to the existing integrated Tensorflow and Apache MXNet deep learning framework containers currently available. Chainer is a popular deep learning framework that supports a variety of neural architectures where the network is defined dynamically in a “define-by-run” scheme. This means you can fully leverage Python constructs and control flows in your network. To quickly get started with Chainer, HAQM SageMaker provides sample notebooks for common workflows such as sentiment analysis and MNIST, by example, easily accessible within the Jupyter Dashboard Interface from HAQM SageMaker.
HAQM SageMaker also now supports AWS CloudFormation, so you can describe and provision all of your infrastructure resources through a template, to standardize configuration across your organization and accounts in an automated and secure manner.
Lastly, HAQM SageMaker is now available in the Asia Pacific (Tokyo) AWS Region.
Chainer integration and support for AWS CloudFormation are available in HAQM SageMaker today in the U.S. East (N. Virginia), U.S. East (Ohio), U.S West (Oregon), EU (Ireland), and Asia Pacific (Tokyo) AWS regions. Visit the documentation for more information on using Chainer in HAQM SageMaker algorithms, and visit the blog post.