Key Feature Additions and Changes | Teradata Vantage API Integration - Key Feature Additions and Changes - Teradata Vantage

Teradata Vantage™ - API Integration Guide for Cloud Machine Learning
Deployment
VantageCloud
VantageCore
Edition
VMware
Enterprise
IntelliFlex
Product
Teradata Vantage
Release Number
1.4
Published
September 2023
ft:locale
en-US
ft:lastEdition
2023-09-28
dita:mapPath
mgu1643999543506.ditamap
dita:ditavalPath
ayr1485454803741.ditaval
dita:id
mgu1643999543506

The following table lists the key feature additions and changes of Teradata Vantage API Integration.

Date Release Feature Additions and Changes
API_Request UDF teradataml Extension (tdapiclient Package)
September 2023 1.04.00.00 Added support for OpenAI and Azure OpenAI Static API_Request UDF function extends support for OpenAI and Azure OpenAI
August 2023 1.03.00.00   Added support for the following:
  • TDApiClient Class to fit and deploy Google Vertex AI models within the Vertex AI environment
July 2023 1.02.01.00   Added support for the following:
  • Static API_Request UDF function to call cloud services/API’s outside of AWS, Azure and Google
  • BYOM support for Sagemaker models through TDApiClient.deploy method
  • CSV support through TDApiClient fit method, eliminating data hop for CSV datasets
  • Explicit requirement for the user to allow data format conversion through the client when JSON format is requested - otherwise an error is thrown
February 2023 1.02.00.00
  • Added support for calling Google Vertex AI endpoint
  
November 2022 1.01.01.00   Added support for the following:
  • TDApiClient Class to fit and deploy Azure Machine Learning models within the Azure Machine Learning environment
  • TDPredictor Class to predict Azure Machine Learning models on Vantage (or through client)
  • Support for transporting CSV and Parquet files to Azure Blob via TDApiClient fit method

    Hops through client is not required.

  • BYOMPredictor Class to support the BYOM feature
July 2022 1.01.00.00
  • Added support for calling Azure Machine Learning endpoint to score model built in Azure Machine Learning
  
May 2022 1.00.00.00
  • Support for calling Amazon SageMaker endpoint to score model built in Amazon SageMaker
 Support the following:
  • TDApiClient Class to fit and deploy Sagemaker models within the Amazon SageMaker environment
  • TDPredictor Class to predict Sagemaker models on Vantage (or through client)
  • Support for transporting Parquet files to S3 via TDApiClient fit method

    Hops through client for other file formats on S3