Monday, 30 July 2018

Amazon SageMaker enables you to process batch jobs for non-Real time inferencing with the new capability of Batch Transform jobs

Amazon SageMaker has added support for High Throughput Batch Transform Jobs for the non-Real time inferencing where the customers can process batch jobs within the Amazon SageMaker irrespective of data set sizes. The Existing machine learning models developed on the Amazon SageMaker can function smoothly with the latest support without making any changes. Earlier to process a batch of data’s for non-real time inferencing the customers had to resize large datasets into smaller chunks and managed real-time endpoints. Now with the latest update, Batch Transform job can be run with any range or size of data sets. SageMaker manages the provisioning of the resources at the start of the job. It removes all the barriers that slow down the performance to use the machine learning model. When the job is complete the output of the batch transform jobs is stored in the S3 bucket. 

No comments:

Post a Comment

AWS Elemental MediaLive annexes Input Switching for Live Channels

AWS Elemental MediaLive is a broadcast-grade live video processing service. It allows you to make high-quality video streams for delivery...