R scripting

The R Scripting action allows you to transfer data to the R environment, run analytical scripts using the data, produce a report in HTML, and populate an existing table with output data from the analysis.

  • Input tables are available as data frames in the R environment.
  • Output tables must be defined in the R script as data frames with a matching schema; i.e., the Data Frame column names must match the column names of the output table.

On the Configuration tab:

  1. Define the R Script with parameters.

  2. Use the Report Options to include R command calls in the exported report.

Possible applications include:

  • Generate descriptive statistics about the dataset
  • Investigate correlations between variables
  • Generate Predictive Analytics models
This action is not supported in the Supply Chain platform.

Run native R script

  1. Drag the R Scripting icon onto the design surface.
  2. On the Connections tab , enter a Name and a Description to identify this action, then specify the database and table:

    Input Tables -

    • For Database Connection, select the database containing one or more tables on which the R Script is to be executed, or choose New Database Connection to establish a new connection.
    • For Table Name, select one or more tables containing the input data. To remove a table from the list of selected tables, highlight the table name, and click the Remove icon .

    Output Tables -

    • For Database Connection, select the database of the output table(s).
    • For Data Frame Name, use the drop-down to select one or more tables for the R Script results. To remove a table from the list of selected tables, highlight the table name, and click the Remove icon .
  3. On the Configuration tab :
    1. Use the R Script Builder to create the R code. You can copy and paste code from another source, or you can type the code directly in the work area of the builder.
      • To add a parameter you have already created, add a left bracket character, and begin typing it in the builder, then select it when it is identified.
      • To view the actual R code created with the builder, click on the R Script tab at the bottom of the action’s window.

    2. Use the checkbox in Report Options to generate a HTML report of the interaction of Data Guru with R. The contents of the report are determined by the Change tool preferences setting in Changing user preferences.

Last modified: Thursday December 19, 2024

Is this useful?