This example focuses on energy demand forecasting, where the goal of a power grid operator is to predict future energy demand given forecasted weather data.
Example: Energy demand forecastingĪutomated machine learning can be used for regression (the prediction of continuous values), classification, or forecasting.
AZURE SQL SERVER CODE
Instructions and code for running the following example are available on GitHub. This allows SQL Server to call Azure ML automated machine learning. Starting in SQL Server 2017, SQL Server includes the ability to run Python code using the sp_execute_external_script stored procedure. Automated machine learning can be used from SQL Server Machine Learning Services, python environments such as Jupyter notebooks and Azure notebooks, Azure Databricks, and Power BI. It chooses the pipelines using its own machine learning model based on the scores from previous pipelines. Automated machine learning tries a variety of machine learning pipelines. We call the service from SQL Server to manage and direct the automated training of machine learning models in SQL Server. Azure Machine Learning serviceĪzure Machine Learning service is a cloud service. This is well suited for use with data residing in SQL Server tables and provides an ideal solution for any version of SQL Server that supports SQL Server Machine Learning Services. While the previous post dealt with a Spark-based implementation tuned for big data, this post presents an approach that runs directly in SQL Server running on a single server. In today’s post, we will present a complementary automated machine learning approach leveraging Azure Machine Learning service (Azure ML) invoked from SQL Server. Recently, we blogged about performing automated machine learning on SQL Server 2019 big data clusters. This post was co-authored by Jeff Shepherd, Deepak Mukunthu, and Vijay Aski.