Apply trained model to new input data

Read data and trained decision tree model

Pre-processing (data preparation)

Transform the data in the same way it was prepared during the training of the model.

How to classify data using a trained model?

Step 1: Drag the "Decision Tree Predictor" node into the workflow and open the configuration window. Connect the output of the "PMML Reader" node to Port 0 and the input data to Port 1.

Step 2: Decide whether you want to change the prediction column name and to append the normalized probabilities to the output.

Step 3: Click "Apply and Execute" to execute the node.

Read a model using a PMML Reader node

Step 1: Drag the "PMML Reader" node into the workflow and click on it to open the configuration window.

Step 2: Use the "Browse" option to locate the model and click "Choose file". Click "Apply" to save the settings.

Step 3: Execute the node to read the model.

Analyze Data: Applying a Churn Predictor

This workflow demonstrates the use case of accessing a trained model and using it for churn prediction on new set of rows. The PMML Reader node is used to read the trained model that was written by the previous workflow in this workflow group.

Link to the training workflow: https://hub.knime.com/s/tqVpSXd1crC_Fes6

Workflow complete!

Keep the momentum going by exploring Just KNIME It! on the Hub to challenge yourself and see how these nodes can be integrated into more complex workflows and use cases.