Predictor workflow that is exposed as RESTful web service

This workflow accesses a pretrained model and applies it to deployment data. The deployment data are provided in JSON format by another workflow/application. The predicted results are sent back in JSON format. Typically an instance of this workflow would be instantiated once (or multiple times to increase performance) and then re-used for new data as it becomes available. The JSON In/Output nodes define the endpoints. Configuration nodes can be used to define additional parameters (not used here).


This is a companion discussion topic for the original entry at https://kni.me/w/Bw2OTMjeBPv3Yque