TechBytes: Using Python with Vantage | 4b. In-Database scripting – Micromodeling and Map Functions
www.teradata.com/vantage
Are you a Data Scientist who love using Python and Jupyter Notebook but having a difficulty building performant analytics and machine-learning at scale? Don’t despair - Teradata Vantage and Teradata Package for Python (teradataml) are to the rescue. They enable performant execution of complex analytics on large datasets, while using your favorite data science tool and language.
In this 4th episode in Using Python with Vantage TechBytes series, Alexander Kolovos demonstrates how to train and score multiple models (micromodeling) in parallel and at scale using Vantage and its SCRIPT Table Operator. It also includes a demonstration for map_row() function for row-based operations and map_partition() function for partition-based operations.
Get a full understanding of the latest features offered by teradataml and how Vantage coupled with teradataml can drive faster time to value through this TechBytes series:
• Part 1. Introduction and Connections [ • TechBytes: Using Python with Vantage ... ]
• Part 2. Data Exploration and Transformations | Building an Analytic Data Set (ADS) [ • TechBytes: Using Python with Vantage ... ]
• Part 3. Analytic Functions Modeling and Model Cataloging [ • TechBytes: Using Python with Vantage ... ]
• Part 4a. In-Database scripting with SCRIPT Table Operator - Scoring with External Model [ • TechBytes: Using Python with Vantage ... ]
• Part 4b. In-Database scripting with SCRIPT Table Operator - Micromodeling (multiple model training and scoring) and Map Functions [ • TechBytes: Using Python with Vantage ... ]
Download Jupyter notebook used in the demonstration from a Teradata GitHub site: https://github.com/Teradata/techbytes...