Workspace Node: Deployment Using PMML - PMML Tab

In the Deployment using PMML node dialog box, select the PMML tab to display the following options.

Element Name Description
Load PMML Click this button to display a drop-down list containing commands to load a PMML file from disk or from Enterprise.

After loading a PMML file, the field below this button shows the PMML file code. The name of the file and the dependent and predictor variables are listed on the Details tab.

Open from disk Displays the Open dialog box, where you can browse to and load a previously saved PMML file.
Open from Enterprise Displays the Select model dialog box, where you can browse to and load a PMML file previously saved in Enterprise.
Copy Select this option  to copy the PMML file from the source (disk or Enterprise). This option is disabled when the node is created as downstream from the model.
Link Select this option to link the PMML file from the source (disk or Enterprise). This option is disabled when the node is created as downstream from the model.
Deploy to Enterprise
    1. Select this button to add/deploy the model as an analysis configuration to Statistica Enterprise. The Choose Deployment Type dialog box displays.
    1. Specify whether you are deploying a new object or updating an existing object.
Deploy to SBDA

This option is available enabled when the PMML node is loaded with Statistica Tree model PMML code.

    1. Click this button to display the Select a Statistica Big Data Analytics Server dialog box.
    1. Select a server.
    1. Click OK.

Statistica checks the validity of the server. The SBDA Deployment dialog box displayed.

Note: When you click OK in the SBDA Deployment dialog box, Statistica internally generates Java MapReduce code from the PMML. The generated Java MapReduce code is appended to the CData section of the CompileStatistica command API. The supplied Name and Overwrite details are also added to this command API. The command API is then POSTED as an HTTPWebClient request from the Statistica side. After the client side POSTing, the web server side of SBDA compiles the Java MapReduce code available in the CData section of the command and creates a new Processing element (with the given name) within the Analysis pane of SBDA.
Options

See Common Options.

OK
  • Click this button to accept all the specifications made in the dialog box and to close it.
  • To view the new spreadsheet, click the icon on the lower-right corner of the Concatenate Variables node.