Importing a PMML model

Apart from creating your own models from Pega Platform™ templates, you can import predictive models in the PMML format created in third-party tools to predict customer actions.

The PMML supported models are:
  • Clustering
  • GeneralRegressionModel
  • MiningModel
  • NaiveBayesModel
  • NearestNeighborModel
  • NeuralNetwork
  • RegressionModel
  • RuleSetModel
  • Scorecard
  • SupportVectorMachineModel
  • TreeModel
  1. In the navigation panel of Prediction Studio, click Predictions.
  2. In the header of the Predictions work area, click New > Predictive model.
  3. In the New predictive model dialog box, enter a Name for your model.
  4. In the Create model section, click Import PMML.
  5. Click Choose and select a model file to upload.
  6. In the Save model section, specify the context where you want to save the model:
    If Then
    If you want to save your model in the default application context, select the Use default context check box.

    For more information, see Configuring the default rule context.
    If you want to save your model in a custom context, specify the context details:
    • Place the cursor in the Apply to class field, press the Down arrow key, and click the class in which you want to save the model.
    • Define the class context by selecting appropriate values from the Development branch, Add to ruleset, and Ruleset version lists.
  7. Verify the settings and click Next.
  8. Optional: In the Outcome definition section, change the default label for the model objective by clicking Set labels and entering a meaningful name in the associated field.
    To enable response capture, the model objective label must be the same as the .pyPrediction parameter value in the response strategy (applies to all model types).
  9. In the Outcome definition section, specify what the model predicts:
    To enable response capture for binary models, the label of the predicted outcome that you want to monitor must be the same as the .pyOutcome parameter value in the response strategy.
    If Then
    If you are importing a binary (scoring) model, perform the following actions:
    1. In the Monitor the probability of field, select the outcome that you want to predict.
    2. In the Advanced section, enter the expected score range.
    3. In the Classification output field, select one of the model outputs to classify the model.
    If you are importing a special binary (scoring) model to predict a number, perform the following actions:
    1. In the Predicting list, select continuous value.
    2. In the Predicting values between fields, enter the range of outcome values that you want to predict.
    If you are importing a categorical (multi-class) model, in the Predicting section, verify the categories to predict.
    If you are importing a continuous model, in the Predicting values between fields, enter the range of outcome values that you want to predict.
  10. In the Expected performance field, enter a value that represents the expected predictive performance of the model.
    The performance measurement metrics are different for each model type. For more information, see Metrics for measuring predictive performance.
  11. Confirm the PMML model settings by clicking Import.
Result: Your custom model is now available in Pega Platform.
What to do next: Configure your PMML model. For more information, see Configuring a PMML model.