ibmdbR (version 1.34.3)

ibmdbR-package: IBM In-Database Analytics

Description

In-database analytics functions operate directly on data in a database, rather than requiring that the data first be extracted to working memory. This lets you analyze large amounts of data that would be impractical or impossible to extract. It also avoids security issues associated with extracting data, and ensures that the data being analyzed is as current as possible. Some functions additionally use lazy loading to load only those parts of the data that are actually required, to further increase efficiency. This package also contains a data structure called a ida.list, which you can use to store R objects in the database. This simplifies the sharing of R objects among users. Each user is assigned two tables for R object storage: a private table, to which only that user has access, and a public table, which can be read by other users. Use a IDA list to generate a pointer to either of these tables, and use the pointer to list, store, or retrieve R objects.

Examples

# Connect to the database and initialize the analytics component.
con <- idaConnect("BLUDB","","")
idaInit(con)

# Create a link to a table IRIS in the database.
iris.ida <- ida.data.frame("IRIS")

# Inspect the data.
iris.ida
head(iris.ida)
dim(iris.ida)
names(iris.ida)

# Create a sub-selection.
iris.ida2 <- iris.ida[iris.ida$Species=='setosa',1:3]
dim(iris.ida2);

# Create a linear model in the database.
linModel <- idaLm(SepalWidth~SepalLength,iris.ida)
linModel

# Create a pointer to the private R object storage table of the current user.
idaList <- ida.list()

# Store the R object that contains the linear model in the 
# private R object storage table of the current user.
idaList[1] <- linModel;

# Retrieve the linear model.
linModelCopy <- idaList[1];
linModelCopy

# Close the connection.
idaClose(con)