Skip to content

Latest commit

 

History

History
119 lines (102 loc) · 5.93 KB

CHANGELOG.md

File metadata and controls

119 lines (102 loc) · 5.93 KB

Change Log

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[1.5.4] - 2023-04-18

Updated

  • The bucket policy on the logging bucket to grant access to the logging service principal (logging.s3.amazonaws.com) for access log delivery.
  • Python libraries.

[1.5.3] - 2023-03-22

Fixed

Updated

  • Python libraries.

[1.5.2] - 2023-01-10

Updated

  • Python libraries.

Removed

  • Use of python future library

[1.5.1] - 2022-12-19

Fixed

[1.5.0] - 2022-12-05

Added

  • Service Catalog AppRegistry resource to register the CloudFormation template and underlying resources as an application in both Service Catalog AppRegistry and AWS Systems Manager Application Manager.

Updated

  • CDK version 1.173.0

[1.4.1] - 2021-08-23

Fixed

  • Upgraded Python package versions

[1.4.0] - 2021-12-22

Added

  • AutoPredictor support has been added.
  • Resources can now be tagged using arbitrary, user-provided tags.

Changed

  • The notebook instance now uses the platform notebook-al2-v1.
  • Upgrading from earlier versions of the stack is not supported. Please redeploy the stack and copy your configuration and data to the newly created forecast data bucket.

[1.3.3] - 2021-06-23

Changed

  • The default notebook instance type is now ml.t3.medium for new stacks to improve performance and availability.

Fixed

  • Fixed an issue that might cause the stack to fail to deploy. This failure occurred when AWS::S3::Bucket ForecastBucket deployed before a required AWS::Lambda::Permission.

[1.3.2] - 2021-06-17

Changed

  • Amazon Athena engine version 2 is now the default requested workgroup version.

[1.3.1] - 2021-04-08

Fixed

  • Upgraded Python package versions and removed unused packages

[1.3.0] - 2021-03-02

Added

  • The solution now supports the Amazon Forecast Weather Index. For usage examples, reference the Implementation Guide.
  • The solution now includes a demo stack (forecasting the number of NYC City taxi pickups in the next 1 hour in each of 260 pickup zones). The demo features AWS managed weather data and holiday features.

Changed

  • The solution now uses an AWS Glue version 2.0 Spark job to transform and aggregate forecast input data and metadata, predictor backtest export data and forecast export data for use in Amazon Athena and Amazon QuickSight for query and visualization.

Fixed

[1.2.0] - 2020-11-23

Added

Fixed

  • The AWS Step Functions State Machine now tags all dataset files uploaded to Amazon Forecast using a full checksum rather than the S3 entity tag. This provides better support for multipart uploads and larger datasets.

[1.1.0] - 2020-10-26

Added

  • Datasets can now be shared between predictor and forecast configurations. As these shared datasets are uploaded, all dependent resources are updated in Amazon Forecast. See the Implementation Guide for details.
  • Synthetic dataset generation tooling has been improved to support metadata and related time series data. See source/synthetic/README.md for details.
  • After forecast exports are completed, the solution now outputs an Athena table of the same name, combining forecast input time series, input metadata, and forecast export into one location. The name of this table matches the name of the forecast export in the Amazon Forecast console. These tables are partitioned for data frequencies greater than daily using YYYY-MM, and will be output in the Forecast Data bucket under the prefix exports/<export_name>
  • After forecast exports and Athena tables are created, the solution now automatically creates an interactive and shareable analysis in Amazon QuickSight with pre-set visuals and configurations to analyze your forecasts. To enable this functionality, provide the QuickSight Analysis Owner ARN parameter in CloudFormation and ensure that QuickSight Edition is enabled in your account and region where this solution is deployed.

Fixed

  • The step function now gracefully handles Amazon Forecast rate limiting. Should you need to run a large number of concurrent experiments (e.g. training more than three predictors/ running more than three forecasts simultaneously), it is recommended that you request a default quota increase with AWS support.

[1.0.1] - 2020-08-07

Fixed

  • fixed a known issue causing multiple dataset upload to fail
  • fixed an issue where users would be notified multiple times that a forecast had completed successfully

Changed

  • anonymous solution metrics publish on all stack changes
  • improved update emails as datasets are uploaded

[1.0.0] - 2020-07-02

Added

  • initial release