Vantage ClearScape Analytics consists of comprehensive in-database data manipulation, modeling and scoring functions, open and connected third-party integrations and APIs, as well as features enabling full-scale activation and operationalization of analytics.
- In-Database Analytics, includes a wide variety and rich set of analytic functions for data preparation, cleaning, transformation, feature engineering, model training and model scoring.
- Bring Your Own Analytics, refers to a highly extensible framework that enables execution of open-source code or libraries and import of pre-trained models for scoring data in Vantage.
In addition, Teradata provides client analytic libraries for Python (teradataml) and R (tdplyr) languages.
- Partner Integration, allows superior user experience by integrating partner analytics closely to the data and parallelizing computations on Vantage.
Teradata provides the following ways to analyze your data on VantageCloud Lake.
- Vantage Analytics Library
Use over 50 advanced analytic functions that support the entire data science process (including in-database predictive modeling, evaluation, and scoring) with the Vantage Analytics Library tool.
- Bring Your Own Model
Use Bring Your Own Model (BYOM) to bring machine learning models, built using popular tools or platforms, and score them at scale in Vantage with no data movement.
- In-Database Analytic Functions
Explore, transform, model and score your data with in-database functions.
- Text Analytics AI Functions
Use a range of text analytics functions using large language models available on various cloud platforms, utilizing data stored in Vantage. This expansion enables execution on the database side using fast path functions utilizing the parallel processing capabilities of Vantage.
- Build Scalable Analytics with Open Analytics Framework
Score Python machine learning models with Open Analytics Framework. See introduction and use cases.
Sample use cases are available for download as attachment OpenAF_Documentation _SampleUseCases.zip. - Teradata Package for Python on VantageCloud Lake
Explore, transform, model and score your data with Python client library - teradataml.
See samples in this section and in the Teradata Package for Python Function Reference on VantageCloud Lake
Sample workflows for teradataml BYOM and Open Analytics Framework features are available for download from the left pane as attachments teradataml_workflows_byom.zip and teradataml_workflows_opaf.zip respectively. - Teradata pyspark2teradataml
teradatamlspk is the Python package name of Teradata product pyspark2teradataml. The teradatamlspk package is built as an extension of teradataml - a Teradata Package for Python. This allows the existing PySpark workloads that runs on Spark engine to easily run on Teradata Vantage using ClearScape Analytics with minimal changes to the PySpark workloads.
- Teradataml Widgets on VantageCloud Lake
Use teradatamlwidgets plug-in components in a notebook to access and execute analytic functions provided by Teradata.
- Teradata Package for Generative AI
The teradatagenai package is a Python library that provides access to various large language models and conducts AI-driven text analytics seamlessly using data from Teradata Vantage.
- Database Unbounded Array Framework Time Series Functions
Time series and forecast with in-database Unbounded Array Framework (UAF) functions.
- API Integration Guide for Cloud Machine Learning
Learn about Teradata API integrations including Teradata Partner, AWS, Azure Machine Learning, Google Vertex AI, and OpenAI, Azure OpenAI, and Amazon Bedrock.
- nPath Visualization
Analyze web site clicks, sensor data, financial data, and more with nPath Visualization.
- Character Shorthand Notation Used in This Document
- How to Read Syntax