Machine Learning Templates with SQL Server 2016 R Services
Discover more examples at Microsoft Machine Learning Server
In these examples, we will demonstrate how to develop and deploy end-to-end advanced analytics solutions with SQL Server 2016 R Services.
Develop models in R IDE. SQL Server 2016 R services allows Data Scientists to develop solutions in an R IDE (such as RStudio, Visual Studio R Tools) with Open Source R or Microsoft R Server, using data residing in SQL Server, and computing done in-database.
Operationalize models in SQL. Once the model development is completed, the model (data processing, feature engineering, training, saved models, and production scoring) can be deployed to SQL Server using T-SQL Stored Procedures, which can be run within SQL environment (such as SQL Server Management Studio) or called by applications to make predictions.
Machine Learning Templates. We have developed a number of templates for solving specific machine learning problems with SQL Server R Services. These templates provides a higher starting point and aims to enable users to quickly build and deploy solutions. Each template includes the following components:
- Predefined data schema applicable to the specific domain
- Domain specific data processing and feature engineering steps
- Preselected *training *algorithms fit to the specific domain
- Domain specific evaluation metrics where applicable
- Prediction (scoring) in production.
The available templates are listed below. Please check back often, as there will be new templates added periodically.
- Predictive Maintenance. Predict machine failures.
- Customer Churn Prediction. Predict when a customer churn happens.
- Online Purchase Fraud Detection. Predict if an online purchase transactions is fraudulent. Versions of this template can be deployed from the Cortana Intelligence Gallery with SQL Server or HDInsight Spark Cluster
- Energy Demand Forecasting. Forecast electricity demand of multiple regions.
- Retail Forecasting Forecast the product sales for a retail store.
- Campaign Management Predict when and how to contact potential customers. Versions of this template can be deployed from the Cortana Intelligence Gallery with SQL Server or HDInsight Spark Cluster
Templates with SQL Server R Services. In these templates, we show the two version of implementations:
- Development Code in R IDE
- Operationalization In SQL
The following is the directory structure for each template:
- Data This contains the provided sample data for each application.
- R This contains the R development code (Microsoft R Server). It runs in R IDE, with computation being done in-database (by setting compute context to SQL Server).
- SQLR This contains the Stored SQL procedures from data processing to model deployment. It runs in SQL environment. A Powershell script is provided to invoke the modeling steps end-to-end.
Performance Turning. This folder provide a few tips on how to improve performance of running R script in SQL Server compute context.
NOTE: Please don't use "Download ZIP" to get this repository, as it will change the line endings in the data files. Use "git clone" to get a local copy of this repository.