This workflow shows a simple example of churn prediction deployment with optimized threshold. This workflow is used for comparison before Integrated Deployment was introduced with KNIME 4.2. You can see how the trained models are manually loaded via Model Reader nodes and the optimized threshold hardcoded via a Rule Engine node.
This is a companion discussion topic for the original entry at https://kni.me/w/86Q-OAKlfE3Za0ue