Does this skill require moving data out of Teradata?

No, the skill generates optimized SQL that executes natively within Teradata Vantage using ClearScape Analytics, ensuring data stays secure and processing remains local.

What permissions are required to use this skill?

You need CREATE, DROP, and SELECT permissions on your working Teradata database, as well as access to the TD_SVM and SVMSparsePredict functions.

What types of classification does this SVM skill support?

It supports both linear and non-linear classification using advanced kernel methods provided by the Teradata TD_SVM function.

Does it handle categorical variables automatically?

Yes, the skill includes logic for automated preprocessing, which identifies categorical columns and generates the necessary encoding SQL.

Teradata SVM Classification

Name: Teradata SVM Classification
Author: teradata-labs

byteradata-labs

0•

Data Science & ML

Automates the end-to-end Support Vector Machine (SVM) workflow for linear and non-linear classification tasks within Teradata Vantage.

This skill streamlines the implementation of sophisticated classification models by leveraging Teradata's native TD_SVM function and ClearScape Analytics. It manages the entire analytical pipeline—from automated feature engineering and data preprocessing to model training and production-ready SQL generation—enabling data scientists to build robust, scalable predictive models directly on their Teradata platform without the need for data movement. It is particularly effective for high-dimensional data and complex non-linear classification problems where performance and data gravity are critical.

Key Features

01Intelligent table analysis for automatic feature and data type detection

02Automated data preprocessing including scaling, encoding, and train-test splitting

03Production-ready SQL generation for end-to-end analytical workflows

04Support for both linear and non-linear SVM kernels

05Comprehensive model evaluation metrics and data quality validation

060 GitHub stars

Use Cases

01Identifying fraudulent transactions using non-linear classification patterns

02Multi-variate classification for predictive maintenance and risk assessment

03Predicting customer churn or retention using high-dimensional behavior data

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add teradata-labs/claude-cookbooks analytics

For use in Claude.ai and ChatGPT

Download Skill