Looking at Retrieve and Rank Web UI bound to the conversation-enhanced application:
https://github.com/watson-developer-cloud/conversation-enhanced
no questions have been uploaded for training, though there is a trainingdata.csv.
I would like to understand how trainingdata.csv was constructed.
Thank you !
That training data was created manually, not using the UI, using the approach described in https://www.ibm.com/watson/developercloud/doc/retrieve-rank/training_data.shtml (because it was prepared before the tooling was available)