Rapid Deployment of Predictive Models - Save Results Tab

Select the Save results tab of the Rapid Deployment of Predictive Models dialog box to access options to save predicted values, classifications, or classification probabilities into an existing data file, into designated variables (that were already defined for and exist in that data file). If your input data file is a Streaming Database Connection file (e.g., on a remote server), ensure that the cursor is properly defined to permit saving (writing) of information to the database.

Data file to write back. Click this button to display a standard file selection dialog box, where you can select a data file to which to write the predictions (predicted classifications, or classification probabilities).

Save predicted & residual values to input data. This option is only available after you designate a data file (or streaming connection to a database) to which to write the predictions. (Note that this file should already contain variables or database fields to hold the predicted values, classifications, and so on.) After clicking this button, the Assign Statistics to Variables for Saving in Input Data dialog box is displayed, where you can specify which statistics (predictions, probabilities, etc.) to save to which specific variables in the data file, or fields in the database. The program then computes the requested prediction statistics, and saves them in the selected variables or fields.

Include prediction probabilities in output. Select this check box when you not only want to save predicted classifications, but also classification probabilities. This option is only available for classification problems, and not for regression-type problems.