td-data-preparation

Validate, format, and impute time series data for Teradata UAF.

7|Updated Dec 3, 2025

Search Tags:#analytics #time series #data validation #etl #data preparation #uaf #teradata

Ready to run

npx skills add https://github.com/teradata-labs/claude-cookbooks --skill td-data-preparation

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: td-data-preparation
Source: https://github.com/teradata-labs/claude-cookbooks/tree/main/skills/analytics/td-data-preparation
Command: npx skills add https://github.com/teradata-labs/claude-cookbooks --skill td-data-preparation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the complex process of preparing raw time series data, ensuring it's clean, validated, and optimally formatted for Teradata's Unbounded Array Framework (UAF) analytics functions.

Core Features & Use Cases

Data Validation: Checks for missing timestamps, duplicate entries, and identifies outliers.
Time Series Formatting: Standardizes temporal data and extracts relevant features (year, month, day, etc.).
Missing Value Imputation: Fills gaps using methods like forward fill or linear interpolation.
Use Case: Prepare sensor data from millions of IoT devices for advanced time series forecasting models within Teradata Vantage.

Quick Start

Prepare the time series data in the 'my_database.sensor_readings' table with timestamp column 'event_time' and value column 'temperature'.

Frequently Asked Questions about td-data-preparation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I prepare time series data for Teradata UAF analytics?▼

To prepare time series data for Teradata UAF, you need to validate timestamps, format temporal features, and impute missing values. This Skill automates that data preparation by detecting outliers and filling gaps using methods like forward fill or linear interpolation.

What is the best way to handle missing values in IoT sensor data before forecasting?▼

Handling missing values in IoT sensor data before forecasting requires imputation to ensure continuity. This Skill fills data gaps using forward fill or linear interpolation methods while simultaneously detecting outliers to maintain dataset integrity.

Can I use this data preparation Skill for large-scale IoT sensor datasets in Teradata Vantage?▼

Yes, you can use this Skill for large-scale IoT sensor datasets in Teradata Vantage. It is specifically designed to analyze and prepare time-dependent data from sources like millions of IoT devices for advanced time series forecasting models.

How do I validate time series data for duplicate entries and missing timestamps?▼

Validating time series data for duplicate entries and missing timestamps is a core feature of this Skill. It checks for missing timestamps, identifies duplicate records, and detects outliers to ensure your dataset is clean and ready for UAF functions.

Does time series formatting for UAF extract features like year, month, and day?▼

Yes, time series formatting for UAF extracts temporal features like year, month, and day. This Skill standardizes your temporal data and extracts relevant time components to optimize the dataset for Unbounded Array Framework analytics functions.

Why does my time series data need validation before using TD_DATA_PREPARATION?▼

Your time series data needs validation before using TD_DATA_PREPARATION because UAF functions require clean, formatted inputs. This Skill ensures data readiness by handling missing values, removing duplicates, and detecting outliers that could skew forecasting results.