The Rapid Deployment node requires a input data node that connect to Rapid Deployment node as input file for prediction. Please refer to this KB article on how to generate advanced I-tree PMML model and deploy it on another dataset through Rapid deployment node. 

For illustration purpose, we load the testing subset of example dataset CreditScoring.sta as input file for Rapid Deployment node. The PMML model is generated from advanced C&RT classification analysis based on the training subset of example dataset CredictScoring.sta (located in C:\Program Files\Statistica\Statistica 13\Examples\Datasets\).

METHOD 1:

1. Double click on the "Rapid Deployment" node, check "Predicted values" and "Save predicted and residual values (classification)". A dialog will be popped out to ask you which variables to be outputted. "Select All" or choose those input variables you want to show in output. Click OK.

2. Run the workspace or the Rapid deployment node. Open the Reporting Document, under "Rapid deployment (beta)" tab, find item "Input spreadsheet generated from deployment". This file shows the prediction result in first two columns and the remaining columns are the input file data that you have chosen to show in output in Step 1. You can click on "Home | Save | Save Item As" to save the spreadsheet to local drive. 

METHOD 2:

1. Go to the input data file for Rapid deployment, create two new variables to store the prediction result, name the new variables, for example, "PredWriteBack" and "PredResWriteBack". 

2. Go to "Rapid Deployment (beta)" node, under "Save Results", check the option "Write back to input file", click on "Assign". In the Assign statistics dialog, assign the two variables created in input data file to store the prediction result. For example, input file variable "PredWriteBack" to store "TreeModelPred" statistics, and "PredResWriteBack" to store "TreeModelRes" statistics. 

3. After assigning variables, click OK and run the Rapid Deployment node. Open the input file node (i.e TestingData"), you can find the prediction statistics were written to the new variables created, namely "PredWriteBack" and "PredResWriteBack".