March 2018
Intermediate to advanced
816 pages
19h 35m
English
Once you have created a model, you can deploy it to a SQL Server table and use it later for predictions. You can also do the predictions in R and store just the results in a SQL Server table.
Let's start by creating another model in R. This time, the model uses the Logistic Regression algorithm. The model uses the SQL Server data and the dbo.vTargetMail view to learn how the values of the NumberCarsOwned, TotalChildren, Age, and YearlyIncome input variables influence the value of the BikeBuyer target variable. The following code sets the execution context back to SQL Server, creates the model with the RevoScale rxLogit() function, and shows the summary of the model:
rxSetComputeContext(srvEx); bbLogR <- rxLogit(BikeBuyer ...