Overview of Using R Models with SAS Model Manager

R is a freely available language and environment for statistical computing and graphics. Using the open architecture of SAS Model Manager, you can register and import R models. SAS Model Manager requires a model template file and model component files that are created specifically for R models.
The following SAS components are required to use R models in SAS Model Manager:
  • Ensure that the installed R language version is 2.13.0 or later.
  • SAS/IML. You must license SAS/IML because the IML procedure is required to export SAS data sets to R and to submit R code.
  • The RLANG system option. You must set this system option.
  • The R language and SAS/IML must be installed on the same machine where the SAS Model Manager server is installed.
SAS Model Manager supplies three R model templates that you can use, or you can create your own template as well. The R model templates that are provided by SAS Model Manager support the analytic, classification, and prediction model functions. The segmentation model function is not supported for R models.
After the model component files are registered, you can perform all SAS Model Manager functions except for exporting an R model to the SAS Metadata Repository.
To use R models in SAS Model Manager, do the following tasks:
  1. Ensure that the RLANG system option is set. To have the RLANG system option set when SAS starts, have your site administrator add the RLANG system option to the SAS configuration file.
  2. Build an R model. For more information, see Build an R Model.
    SAS/IML must be installed before you build an R model.
  3. Ensure that you have a model template file. For more information, see Prepare an R Model Template File.
  4. Ensure that you have the required model component files. For more information, see Prepare R Model Component Files.
  5. Import the R model. For more information, see Import Models from Local Files.
Last updated: June 12, 2017