From data ingestion to model deployment, Teradata empowers data scientists and machine learning engineers to build end-to-end AI/ML pipelines using ClearScape Analytics. In this video, we walk you through the development of a machine learning pipeline in Teradata Vantage, demonstrating how to forecast energy consumption with advanced AI/ML tools and techniques. Accurately predicting energy demand is crucial for energy trading companies to avoid losses, optimize cash flow, and meet regulatory requirements. We compare two machine learning models—Linear Regression and Random Forest—to determine the most accurate predictor of energy consumption.
You’ll learn how to apply Teradata’s in-database SQL functions for efficient data exploration, preparation, training, and scoring. We also highlight the flexibility of training models using external partner environments including AWS SageMaker, Google Vertex AI, Azure ML, DataRobot, Dataiku, and others. With Teradata’s Bring Your Own Model (BYOM) functionality, you can easily load pre-trained models onto Vantage, supporting formats like PMML, ONNX, H2O Mojo, and others.
This video will help you understand how to:
Develop and implement a complete AI/ML pipeline using Teradata Vantage and ClearScape Analytics.
Use in-database scoring for improved efficiency and security.
Seamlessly integrate and deploy models at scale using BYOM – Bring Your Own Model functionality.
Compare machine learning models like Linear Regression and Random Forest for better predictions.
Chapters:
Introduction – 00:00
ClearScape Analytics Experience environment setup – 01:08
Data ingestion, exploration and visualization –2:30
Data preparation with ClearScape Analytics in-database functions– 06:30
Model training – 8:55
Model scoring and evaluation with PMML predict function – 11:06
Visualize and compare predictive results – 12:32