Announce Feathr 1.0.0

Feathr 1.0.0 is released on February 28, 2023, with following new features:

New Features

• Feathr sandbox: The Feathr sandbox is a pre-configured environment that you can use to learn how to use Feathr, experiment with features, and build proof-of-concept applications locally, without setting up complex infrastructure on the cloud. Please check out Quick Start Guide with Local Sandbox to get started.

• Online Transform: Online Transform is a feature that enables real-time feature transformations as part of the feature retrieval process, here are some use cases examples:

  • Featurization source is only available at inference time.
  • Using offline transform might be a waste of storage and compute resources.
  • Users would like to decouple featurization work off up-stream online system.
  • Users would like to define transformation once for both online and offline consumption.  Please checkout feathr-ai/feathr-online (github.com) to get started

• Use Azure Cosmos DB as online store. Please see following documentation for provisioning and materialization

  • Provisioning: Azure Resource Provisioning through Azure Resource Manager

  • Materialization: Using SQL databases, CosmosDb with Feathr

• Feathr Notebook samples now provide 5 getting started tutorial with Jupyter notebooks to help users get started, these notebooks cover various use cases commonly used in ML work.

  Name	Description	Platform
  NYC Taxi Demo	Quickstart notebook that showcases how to define, materialize, and register features with NYC taxi-fare prediction sample data.	Azure Synapse, Databricks, Local Spark
  Databricks Quickstart NYC Taxi Demo	Quickstart Databricks notebook with NYC taxi-fare prediction sample data.	Databricks
  Feature Embedding	Feathr UDF example showing how to define and use feature embedding with a pre-trained Transformer model and hotel review sample data.	Databricks
  Fraud Detection Demo	An example to demonstrate Feature Store using multiple data sources such as user account and transaction data.	Azure Synapse, Databricks, Local Spark
  Product Recommendation Demo	Feathr Feature Store example notebook with a product recommendation scenario	Azure Synapse, Databricks, Local Spark

• Web UI has been enhanced in following areas

  • Support register feature
  • Support delete feature
  • Support display version
  • Lineage has been added to top menu for quick access You can try these new changes on Feathr live demo site at Feathr Feature Store (feathr-registry-sql.azurewebsites.net)

• Use SparkSQL as DataSource, code sample can be found at this file

• Time pattern support in data source path. Please check Input File for Feathr for more details.

• Feature names conflicts check and auto correction, please check Getting Offline Features using Feature Query for more details.

• Feathr MLOps V2 integration with Terraform Deployment. Please see Feathr MLOps V2 Integration Deployment Guide for more details.

For more information, please refer to the official Feathr documentation.