Running Datasets Against Saved Models

How to:

After you save a predictive model, you can load it and run other datasets against it, in order to learn more about your data.

The following binary classification model was previously trained and saved. The model includes a Customer Status column that shows values of Active or Churn, as shown in the following image.

Note: This model is selected in Step 7 below.

You may want to run a different customer dataset against this predictive model. For example, if you want to know which customers will remain active and which will churn. A dataset that does not include a customer status column is shown in the following image.

Note: This dataset is selected in Step 2 below.

For more information about training and saving models, see Running Predictive Analytics On Your Data.

Procedure: How to Run Datasets Against Saved Models

  1. From the WebFOCUS Hub, click the plus menu, and then click Create Data Flow.

    The TIBCO WebFOCUS Reporting Server browser interface opens.

  2. Drag a dataset onto the canvas.
  3. From the left panel, click Functions.
  4. Click the Machine Learning (Python-based) function folder, as shown in the following image.
  5. Click the RUN_MODEL function, as shown in the following image.

    Note: There are two RUN_MODEL functions. RUN_MODEL loads and runs the ML Model. RUN_MODEL2 loads and runs the ML Model and overrides predictors.

    The Properties dialog box opens, as shown in the following image.

  6. In the Python_MFD Parameters field, click the file icon.

    The Model Function dialog opens.

  7. Select the model to be loaded and run, as shown in the following image.
  8. Click OK.

    The Properties dialog box is updated with your selected model.

  9. Click OK.

    The model Metadata dialog opens, as shown in the following image.

    Note: The dataset now includes a Customer Status Predicted column.

  10. Click OK.

    You are returned to the data flow, as shown in the following image.

    The dataset now predicts which customers will remain active and which will churn, as determined by the predictive model, based on values in the dataset.

  11. From the WebFOCUS toolbar menu, select Save As to save your updated dataset.