%mdlmgr_AddProject Macro

Adds a project to a folder.

Note: When you use the %mdlmgr_AddProject( ) macro to create a new project, an initial version is automatically created within the new project. The default version is 1.0. To add a new version to an existing project, use the %mdlmgr_AddVersion( ) macro.

Syntax

Required Arguments

ParentId=parent-UUID-or-path

specifies the UUID of the parent folder or the model repository path for the parent folder.

You can use &folder-Id-variable that is set for the NewFolderId= argument in the %mdlmgr_AddFolder( ) macro as the value of parent-UUID-or-path.
The model repository path is in this form:
//ModelManagerDefaultRepo/MMRoot/folder-name/

Name=project-name

specified the name of the project. The name can contain letters, spaces, the underscore ( _ ), the hyphen ( - ), and the period ( . ).

ModelFunction=model-function

specifies the project model function type. These are the valid values:

  • Classification
  • Prediction
  • Segmentation
  • Analytical
Default Classification

NewProjectId=project-Id-variable

specifies a variable or a macro variable that is used to identify the new project.

SAS Decision Manager creates a global macro variable, %project-Id-variable whose value is the project UUID or the path in the SAS Metadata Repository. You can use &project-Id-variable as the value of a ParentID= argument in the %mdlmgr_AddVersion( ) macro or the FolderId= argument in the %mdlmgr_SetProperty( ) macro. For example, if you set NewProjectId=projectId, you can use ParentId=&projectId in the %mdlmgr_AddVersion( ) macro.
The SAS Decision Manager path is in this form:
 //ModelManagerDefaultRepo/MMRoot/folder-name/project-name

Optional Arguments

Desc=description

specifies a description of the project.

InputVarTable=project-input-variable-table

specifies a data set that must include the input variables that are used by the champion model. If you have several candidate models for your project, make sure that all candidate model input variables are included in the project input table. The data set does not need to contain data. If you use the train table as a project input table, be sure to exclude the target variable.

The input variable table is used to create the inputvar.xml file, which describes all of the model input variables.
Requirement The data set must be a local or network file. This macro does not support project input tables in the SAS Metadata Repository.
Tip The project input table can be defined after the project is created. It must be defined before the project champion model is set.
See Create a Project Input Table

OutputVarTable=project-output-variable-table

specifies a data set that includes only output variables that are created or modified by the champion model. If you have several candidate models for your project, you must make sure that all project output variables are mapped to the champion model output variables. If you use the train table as the project output table, use the SET statement to specify the training table, and use the KEEP statement to specify the variables from the training table that you want in the project output table.

The output variable table is used to create the outputvar.xml file, which describes all of the model output variables.
Requirement The data set must be a local or network file. This macro does not support project output tables in the SAS Metadata Repository.
Tip The project output table can be defined after the project is created. It must be defined before the project champion model is set.
See Create a Project Output Table

Trace=On | Off

specifies whether to supply verbose trace messages to the SAS log.

Default Off
Last updated: February 22, 2017