How to build and deploy a demand forecasting solution with BigQuery ML

Опубликовано 27 января 2021, 17:05

Learn how to use BigQuery ML to train and deploy a demand forecasting solution.

With potentially millions of products, for a data science and engineering team to create multi-millions of forecasts is one thing, but to procure and manage the infrastructure to handle an end-to-end model training and forecasting solution, this can quickly become overwhelming, especially for large businesses.

This is a step-by-step video that explores the demand forecasting pattern in the Notebook linked below and helps walk you through the entire process of building such a system in your organization.

You will learn how to:
•Prepare the training data in BigQuery
•Train and evaluate a time-series model with BigQuery ML
•Visualize the forecasts in a dashboard
•Schedule and automate model retraining

Solutions guide → goo.gle/3om3dcz
Notebook here → goo.gle/39kf9Hc
More Smart Analytics Reference Patterns → goo.gle/3sZnU1m

Subscribe to Google Developers → goo.gle/developers