I built an email channel to receive emails and categorize emails in topics through an NLP model inside Prediction Studio. This model was just deployed to Production and are receiving email requests and text analyzing to categorize the results.
The question is: how can I get those predicted data that was read by my model in Production environment? Is there a way to get the text that was read and the categorization result provided by the model?
I need to get this data to give feedback and retrain the model inside Development environment.
The best approach is to check “Enable Training data“ in the behavior tab during the initial days of Production. This will enable to understand the utterances which are not matching and helps you can download from the training data tab and train in your development environment and migrate the changes to PROD.
This option is already checked in Production environment. But my repository in production environment for Prediction Studio is defaultstore, I don’t know if this is impacting my application. Do I need to change this repository? If yes, can I change it to pegacloudfilestorage?