Training custom pre trained entities using ner_crf

Hello everyone,

I can’t find any information about how to create our own pre trained entities anywhere. Let’s say for example that I am making a movie/tv bot that should be able to recognise movies and television shows. I have a lot of data for this bot but how would I incorporate that into rasa-nlu training? For now, our team is working on writing some scripts that would take the movies data and generate nlu data to train rasa-nlu with. The problem with this approach is that the movies entity will need to retrained everytime some changes are made. Is there a better approach than this if we want to use ner_crf for custom entities?

You can try training the spacy model with your own data. I looking at the same thing, as want to add custom pre trained entities.