Skip to content
This repository was archived by the owner on Jul 10, 2023. It is now read-only.

Releases: awslabs/predictive-segmentation-using-amazon-pinpoint-and-amazon-sagemaker

v1.1.0

17 Dec 17:21
@Ryanjlowe Ryanjlowe
v1.1.0
482bb42
Compare
Choose a tag to compare
v1.1.0 Latest
Latest

Changed

  • Removed code to manage the Pinpoint Event Stream, Kinesis Firehose, and Glue Database configuration.
  • Added dependency on the Digital User Engagement Events Database to manage the Pinpoint Event Stream and event database creation.
  • Modified logic to write a ChurnPrediction Metric to the Endpoints allowing for a Dynamic Predicted to Churn segment instead of a static imported segment.
  • Providing a Parquet formatted file for the sample pinpoint events to match the database configuration.
  • Added manual schema registration in Glue for the Pinpoint endpoint exports, removing dependency on a Glue crawler.
  • Removed static customer data file and moved attributes for ML predictions to endpoints with the creation of the SampleImport.csv file.
  • SageMaker notebook instance placed inside of a VPC for enhanced security of customer data files.
  • Updated SageMaker notebook to use the latest version of the SageMaker Python SDK.

Refer to changelog for more information.

Assets 2
Loading