Analytics Capabilities | VantageCloud Lake - Analytics Capabilities - Teradata Vantage

Teradata® VantageCloud Lake

Deployment
VantageCloud
Edition
Lake
Product
Teradata Vantage
Published
January 2023
Language
English (United States)
Last Update
2024-02-17
dita:mapPath
phg1621910019905.ditamap
dita:ditavalPath
pny1626732985837.ditaval
dita:id
phg1621910019905

VantageCloud Lake offers a wide range of ClearScape Analytics capabilities, namely, native in-database functions, Python and R analytics in-database and client libraries and integrated third-party analytics (Bring your own Model, Cloud Analytics Integration and Partner Product Integration) ClearScape Analytics also offers the best-of-breed choices and the ability to conduct analytics using tools and languages preferred by analytics professionals - SQL, Python, R and third-party tools like Dataiku and H2O. All these together provide organizations with choices that fit their users need.

  • Vantage native SQL in-database analytics provides users a variety of functions for data preparation, data transformation, feature engineering, hypothesis testing, model training and model scoring, among others. It accelerates analytic workflows by providing a highly performant, scalable, enterprise data platform that runs production advanced analytics with little to no data movement to and from other platforms. Data scientists can also build end-to-end machine learning pipelines directly inside Vantage. This is enabled by the availability of popular machine learning algorithms as well as functionality to create pipeline by chaining all functions. The data scientist will be able to operationalize analytics rapidly at scale as well as minimize data movement.
  • Open Analytics Framework, only available in VantageCloud Lake, offers users the ability to install and utilize open source Python and R language analytic libraries to execute analytics in Vantage and benefit from VantageCloud Lake’s performance and auto-scaling capabilities. Users can use in-database functions to prepare and transform data in Vantage, train a model using open source Python or R analytic libraries externally, like scikit-learn, and subsequently score them in Vantage.
  • Vantage Partner Integration has reached a whole new level with its expanding partners ecosystem. Vantage Bring your own Model (BYOM) capability accepts PMML, ONNX, H2O MOJO and Dataiku formatted models to score in Vantage. BYOM conveniently solves the common problem of production deployment. In addition, Teradata provides a client plugin to Dataiku DSS. Data scientists can use Dataiku’s no-code or low-code interfaces to call VantageCloud Lake analytic functions.
  • VantageCloud Lake implements its analytic capabilities as SQL functions. It adds Python and R interfaces on top of that within the client analytic libraries. The client analytic libraries generate and run appropriate SQL queries transparently for the analytic function. Data scientists unfamiliar with SQL can continue to use Python and R languages to access all the ClearScape Analytics capabilities.
Capability Native in-database Python, R in-database BYOM
Data exploration N/A* N/A*
Data preparation N/A* N/A*
Data transformation N/A* N/A*
Feature engineering N/A* N/A*
Hypothesis testing N/A* N/A*
Model training Under development N/A
Model scoring/prediction
Production deployment

* indicates Vantage native in-database functions are recommended for this task.